Search

A major concern in data-driven deep learning (DL) is how to maximize the capability of a model for limited datasets. The lack of high-performance datasets limits intelligent agriculture development. Recent studies have shown that image enhancement techniques can alleviate the limitations of datasets on model performance. Existing image enhancement algorithms mainly perform in the same category and generate highly correlated samples. Directly using authentic images to expand the dataset, the environmental noise in the image will seriously affect the model's accuracy. Hence, this paper designs an automatic leaf segmentation algorithm (AISG) based on the EISeg segmentation method, separating the leaf information with disease spot characteristics from the background noise in the picture. This algorithm enhances the network model's ability to extract disease features. In addition, the Cycle-GAN network is used for minor sample data enhancement to realize cross-category image transformation. Then, MobileNet was trained by transfer learning on an enhanced dataset. The experimental results reveal that the proposed method achieves a classification accuracy of 98.61% for the ten types of tomato diseases, surpassing the performance of other existing methods. Our method is beneficial in solving the problems of low accuracy and insufficient training data in tomato disease detection. This method can also provide a reference for the detection of other types of plant diseases.

Functional electrical stimulation (FES) has been demonstrated as a viable method for addressing motor dysfunction in individuals affected by stroke, spinal cord injury, and other etiologies. By eliciting muscle contractions to facilitate joint movements, FES plays a crucial role in fostering the restoration of motor function compromised nervous system. In response to the challenge of muscle fatigue associated with conventional FES protocols, a novel biofeedback electrical stimulator incorporating multi-motor tasks and predictive control algorithms has been developed to enable adaptive modulation of stimulation parameters. The study initially establishes a Hammerstein model for the stimulated muscle group, representing a time-varying relationship between the stimulation pulse width and the root mean square (RMS) of the surface electromyography (sEMG). An online parameter identification algorithm utilizing recursive least squares is employed to estimate the time-varying parameters of the Hammerstein model. Predictive control is then implemented through feedback corrections based on the comparison between predicted and actual outputs, guided by an optimization objective function. The integration of predictive control and roll optimization enables closed-loop control of muscle stimulation. The motor training tasks of elbow flexion and extension, wrist flexion and extension, and five-finger grasping were selected for experimental validation. The results indicate that the model parameters were accurately identified, with a RMS error of 3.83 % between actual and predicted values. Furthermore, the predictive control algorithm, based on the motor tasks, effectively adjusted the stimulus parameters to ensure that the stimulated muscle groups can achieve the desired sEMG characteristic trajectory. The biofeedback electrical stimulator that was developed has the potential to assist patients experiencing motor dysfunction in achieving the appropriate joint movements. This research provides a foundation for a novel intelligent electrical stimulation model., Competing Interests: Declaration of competing interest The authors declare that they have no competing interests., (Copyright © 2024. Published by Elsevier Ltd.)

Colon adenocarcinoma (COAD), a frequently encountered and highly lethal malignancy of the digestive system, has been the focus of intensive research regarding its prognosis. The intricate immune microenvironment plays a pivotal role in the pathological progression of COAD; nevertheless, the underlying molecular mechanisms remain incompletely understood. This study aims to explore the immune gene expression patterns in COAD, construct a robust prognostic model, and delve into the molecular mechanisms and potential therapeutic targets for COAD liver metastasis, thereby providing critical support for individualized treatment strategies and prognostic evaluation. Initially, we curated a comprehensive dataset by screening 2600 immune-related genes (IRGs) from the ImmPort and InnateDB databases, successfully obtaining a rich data resource. Subsequently, the COAD patient cohort was classified using the non-negative matrix factorization (NMF) algorithm, enabling accurate categorization. Continuing on, utilizing the weighted gene co-expression network analysis (WGCNA) method, we analyzed the top 5000 genes with the smallest p-values among the differentially expressed genes (DEGs) between immune subtypes. Through this rigorous screening process, we identified the gene modules with the strongest correlation to the COAD subpopulation, and the intersection of genes in these modules with DEGs (COAD vs COAD vs Normal colon tissue) is referred to as Differentially Expressed Immune Genes Associated with COAD (DEIGRC). Employing diverse bioinformatics methodologies, we successfully developed a prognostic model (DPM) consisting of six genes derived from the DEIGRC, which was further validated across multiple independent datasets. Not only does this predictive model accurately forecast the prognosis of COAD patients, but it also provides valuable insights for formulating personalized treatment regimens. Within the constructed DPM, we observed a downregulation of CALB2 expression levels in COAD tissues, whereas NOXA1, KDF1, LARS2, GSR, and TIMP1 exhibited upregulated expression levels. These genes likely play indispensable roles in the initiation and progression of COAD and thus represent potential therapeutic targets for patient management. Furthermore, our investigation into the molecular mechanisms and therapeutic targets for COAD liver metastasis revealed associations with relevant processes such as fat digestion and absorption, cancer gene protein polysaccharides, and nitrogen metabolism. Consequently, genes including CAV1, ANXA1, CPS1, EDNRA, and GC emerge as promising candidates as therapeutic targets for COAD liver metastasis, thereby providing crucial insights for future clinical practices and drug development. In summary, this study uncovers the immune gene expression patterns in COAD, establishes a robust prognostic model, and elucidates the molecular mechanisms and potential therapeutic targets for COAD liver metastasis, thereby possessing significant theoretical and clinical implications. These findings are anticipated to offer substantial support for both the treatment and prognosis management of COAD patients.

Purpose: Stereo matching is a crucial technology in the binocular laparoscopic-based surgical navigation systems. In recent years, neural networks have been widely applied to stereo matching and demonstrated outstanding performance. however, this method heavily relies on manual feature engineering meaning that professionals must be involved in the feature extraction and matching. This process is both time-consuming and demands specific expertise., Methods: This paper introduces a novel stereo matching framework DCStereo that realizes a fully automatic neural architecture design for the stereo matching of binocular laparoscopic images. The proposed framework utilizes a densely connected search space which enables a more flexible and diverse architecture composition. Furthermore, the proposed algorithm leverages the channel and path sampling strategies to reduce memory consumption during searching., Results: Empirically, our searched DCStereo on the SCARED training dataset achieves a mean absolute error of 3.589 mm on the test dataset, which outperforms hand-crafted stereo matching methods and other approaches. Furthermore, when directly testing on the SERV-CT dataset, our DCStereo demonstrates better generalization ability than other methods., Conclusion: Our proposed approach leverages the neural architecture search technique and a densely connected search space for automatic neural architecture design in stereo matching of binocular laparoscopic images. Our method delivers advanced performance on the SCARED dataset and promising results on the SERV-CT dataset. These findings demonstrate the potential of our approach for improving clinical surgical navigation systems., (© 2024. CARS.)

Retrograde intrarenal surgery (RIRS) is a widely utilized diagnostic and therapeutic tool for multiple upper urinary tract pathologies. The image-guided navigation system can assist the surgeon to perform precise surgery by providing the relative position between the lesion and the instrument after the intraoperative image is registered with the preoperative model. However, due to the structural complexity and diversity of multi-branched organs such as kidneys, bronchi, etc., the consistency of the intensity distribution of virtual and real images will be challenged, which makes the classical pure intensity registration method prone to bias and random results in a wide search domain. In this paper, we propose a structural feature similarity-based method combined with a semantic style transfer network, which significantly improves the registration accuracy when the initial state deviation is obvious. Furthermore, multi-view constraints are introduced to compensate for the collapse of spatial depth information and improve the robustness of the algorithm. Experimental studies were conducted on two models generated from patient data to evaluate the performance of the method and competing algorithms. The proposed method obtains mean target error (mTRE) of 0.971 ± 0.585 mm and 1.266 ± 0.416 mm respectively, with better accuracy and robustness overall. Experimental results demonstrate that the proposed method has the potential to be applied to RIRS and extended to other organs with similar structures., (© 2023 The Authors. Journal of Applied Clinical Medical Physics published by Wiley Periodicals, LLC on behalf of The American Association of Physicists in Medicine.)

Feature selection aims to reduce the number of features and improve the classification accuracy, which is an essential step in many real-world problems. Multiple feature subsets with different features selected can achieve similar or even the same objective values (e.g., maximize the classification accuracy and minimize the number of selected features). This means the optimal feature subsets of a classification problem may not be unique. However, most existing feature selection methods do not take into consideration finding multiple optimal feature subsets. In this article, a multiobjective differential evolution approach is developed to search for multiple optimal feature subsets. The contributions are three-fold. First, to provide a good starting point, an initialization method considering feature relevance is proposed. Second, a clustering method is used to divide the whole population into multiple subpopulations. In each of these subpopulations, a subarchive utilizes a developed crowding distance to ensure diversity by considering both the search space and the objective space. Finally, the nondominated solutions from all the subarchives are retained in another archive to guide the evolutionary feature selection process, together with an improved hypervolume contribution indicator. The experiments on 14 datasets of varying difficulty show that the proposed approach can evolve a better Pareto front of feature subsets compared with seven other state-of-the-art methods as well as find different feature subsets with similar or the same classification performance.

Aim: To identify predictors of basilar invagination (BI) prognosis and compare diagnostic properties between logistic modeling and machine learning methods., Material and Methods: We conducted a single-center retrospective study. Patients at our hospital who met the inclusion and exclusion criteria were identified between August 2015 and August 2020 for inclusion. Candidate predictors, such as demographics, clinical scores, radiographic parameters, and outcome, were included. The primary outcome was the prognosis evaluated by the change in patient-reported Japanese orthopaedic association (PRO-JOA) score. Conventional logistic regression models and machine learning algorithms were implemented. Models were compared, considering the area under the curve (AUC), sensitivity, specificity, positive and negative predictive values, and calibration curve., Results: Overall, the machine learning algorithms and traditional logistic regression models performed similarly. The postoperative cervicomedullary angle, head-neck flexion angle (HNFA), atlantodental interval, postoperative clivo-axial angle, age, postoperative clivus slope, postoperative cranial incidence, weight, postoperative HNFA, and postoperative Boogaard's angle (BoA) were identified as important predictors for BI prognosis. Among the surveyed radiographic parameters, postoperative BoA was the most important predictor of BI prognosis. In the validation dataset, the bagged trees model performed best (AUC, 0.90)., Conclusion: Through machine learning, we have demonstrated predictors of BI prognosis. Machine learning methods did not provide too many advantages over logistic regression in predicting BI prognosis but remain promising.

Text based Visual Question Answering (TextVQA) is a recently raised challenge requiring models to read text in images and answer natural language questions by jointly reasoning over the question, textual information and visual content. Introduction of this new modality - Optical Character Recognition (OCR) tokens ushers in demanding reasoning requirements. Most of the state-of-the-art (SoTA) VQA methods fail when answer these questions because of three reasons: (1) poor text reading ability; (2) lack of textual-visual reasoning capacity; and (3) choosing discriminative answering mechanism over generative couterpart (although this has been further addressed by M4C). In this paper, we propose an end-to-end structured multimodal attention (SMA) neural network to mainly solve the first two issues above. SMA first uses a structural graph representation to encode the object-object, object-text and text-text relationships appearing in the image, and then designs a multimodal graph attention network to reason over it. Finally, the outputs from the above modules are processed by a global-local attentional answering module to produce an answer splicing together tokens from both OCR and general vocabulary iteratively by following M4C. Our proposed model outperforms the SoTA models on TextVQA dataset and two tasks of ST-VQA dataset among all models except pre-training based TAP. Demonstrating strong reasoning ability, it also won first place in TextVQA Challenge 2020. We extensively test different OCR methods on several reasoning models and investigate the impact of gradually increased OCR performance on TextVQA benchmark. With better OCR results, different models share dramatic improvement over the VQA accuracy, but our model benefits most blessed by strong textual-visual reasoning ability. To grant our method an upper bound and make a fair testing base available for further works, we also provide human-annotated ground-truth OCR annotations for the TextVQA dataset, which were not given in the original release. The code and ground-truth OCR annotations for the TextVQA dataset are available at https://github.com/ChenyuGAO-CS/SMA.

Semantic scene completion (SSC) is a computer vision task aiming to simultaneously infer the occupancy and semantic labels for each voxel in a scene from partial information consisting of a depth image and/or a RGB image. As a voxel-wise labeling task, the key for SSC is how to effectively model the visual and geometrical variations to complete the scene. To this end, we propose the Anisotropic Network (AIC-Net), with novel convolutional modules that can model varying anisotropic receptive fields voxel-wisely in a computationally efficient manner. The basic idea to achieve such anisotropy is to decompose 3D convolution into three consecutive dimensional convolutions, and determine the dimension-wise kernels on the fly. One module, termed kernel-selection anisotropic (KSA) convolution, adaptively selects the optimal kernel sizes for each dimensional convolution from a set of candidate kernels, and the other module, termed kernel-modulation anisotropic (KMA) convolution, directly modulates a single convolutional kernel for each dimension to derive more flexible receptive field. By stacking multiple such anisotropic modules, the 3D context modeling capability and flexibility can be further enhanced. Moreover, we present a new end-to-end trainable framework to approach the SSC task avoiding the expensive TSDF pre-processing as in many existing methods. Extensive experiments on SSC benchmarks show the advantage of the proposed methods.

The existing multi-manipulator sorting method for gangue that utilizes a multi-task allocation strategy is not satisfactory. The single manipulator working space is fixed, lowering the cooperation degree between the manipulators and leading to a low sorting rate. Therefore, this paper proposes a multi-manipulator cooperative sorting method that can work globally. First, a benefit function based on the sorting time and quality of the gangue is constructed by combining the gangue flow information and the manipulator state. The time parameter is obtained via the manipulator's dynamic target tracking trajectory planning algorithm based on PID control. Secondly, the benefits matrix is standardized and updated many times to improve the Hungarian algorithm to achieve task allocation, and the initial solution with priority is obtained. Finally, the solutions are analyzed and processed cooperatively in order of priority. The conflicts between multiple robotic arms are eliminated through task cooperation and trajectory cooperation until the sorting task that the robot arm can execute is obtained from the allocation results. Experiments involving different sorting methods were completed on a multi-arm coal and gangue sorting experimental robot platform. The experimental results show that the sorting efficiency of the proposed method is about 10% and 20% higher than that of the fixed space dynamic and designated space fixed points methods, respectively, under different belt speeds. This method can guarantee system benefits, effectively implements cooperative control of multi-manipulator operations in the whole area, and improves the efficiency of coal gangue sorting., Competing Interests: The authors declare no conflicts of interest.

Objectives: To reduce the dimension of characteristic information extracted from pelvic CT images by using principal component analysis (PCA) and partial least squares (PLS) methods. To establish a support vector machine (SVM) classification and identification model to identify if there is pelvic injury by the reduced dimension data and evaluate the feasibility of its application., Methods: Eighty percent of 146 normal and injured pelvic CT images were randomly selected as training set for model fitting, and the remaining 20% was used as testing set to verify the accuracy of the test, respectively. Through CT image input, preprocessing, feature extraction, feature information dimension reduction, feature selection, parameter selection, model establishment and model comparison, a discriminative model of pelvic injury was established., Results: The PLS dimension reduction method was better than the PCA method and the SVM model was better than the naive Bayesian classifier (NBC) model. The accuracy of the modeling set, leave-one-out cross validation and testing set of the SVM classification model based on 12 PLS factors was 100%, 100% and 93.33%, respectively., Conclusions: In the evaluation of pelvic injury, the pelvic injury data mining model based on CT images reaches high accuracy, which lays a foundation for automatic and rapid identification of pelvic injuries.

The embryonic stem cell (ESC) has the capacity to self-renew and maintain pluripotent, while continuously offering a source of various differentiated cell types. The fate decision process of remaining in the ground state or transiting to a differentiated state can be read out by the regulatory network of key transcription factors (TFs). However, its underlying mechanism remains to be fully elucidated. In this paper, we tackle this problem by proposing a novel cellular differentiation model for mouse embryonic stem cell (MESC) dynamics regulation: MESC-DRM. We employ nonlinear least-squares algorithm to infer model parameters by using benchmark datasets, construct a potential function by exploiting multivariate Gaussian distributions, and project the potential landscape into a 3D space to validate and replicate the stable cell states observed in experiments. The traditional cell landscape modeling techniques rely on the potential function visualization to decide the stable states of cells. But the visualization will be almost impossible when the dimensionality of the potential function is greater than 3. We handle the challenge by innovatively employing a Lyapunov method to resolve it through a more straightforward analytical approach. It also provides a more rigorous and robust way for accurate cell fate decision. The study not only validates the previous experimental results but also provides an insightful guide for cell fate decision besides inspiring future study on this topic., (Copyright © 2022. Published by Elsevier Ltd.)

The camera is the main sensor of vison-based human activity recognition, and its high-precision calibration of distortion is an important prerequisite of the task. Current studies have shown that multi-parameter model methods achieve higher accuracy than traditional methods in the process of camera calibration. However, these methods need hundreds or even thousands of images to optimize the camera model, which limits their practical use. Here, we propose a novel point-to-point camera distortion calibration method that requires only dozens of images to get a dense distortion rectification map. We have designed an objective function based on deformation between the original images and the projection of reference images, which can eliminate the effect of distortion when optimizing camera parameters. Dense features between the original images and the projection of the reference images are calculated by digital image correlation (DIC). Experiments indicate that our method obtains a comparable result with the multi-parameter model method using a large number of pictures, and contributes a 28.5% improvement to the reprojection error over the polynomial distortion model.

Maintaining a balance between convergence and diversity is particularly crucial in evolutionary multiobjective optimization. Recently, a novel dominance relation called "strengthened dominance relation" (SDR) is proposed, which outperforms the existing dominance relations in balancing convergence and diversity. In this article, two points that influence the performance of SDR are studied and a new dominance relation, which is mainly based on SDR, is proposed (CSDR). An adaptation strategy is presented to dynamically adjust the dominance relation according to the current generation number. The CSDR is embedded into NSGA-II to substitute the Pareto dominance, labeled as NSGA-II/CSDR. The performance of our proposed method is validated by comparing it with five state-of-the-art algorithms on commonly used benchmark problems. NSGA-II/CSDR outperforms other algorithms in the most test instances considering both convergence and diversity.

Skeleton-based action recognition is widely used in varied areas, e.g., surveillance and human-machine interaction. Existing models are mainly learned in a supervised manner, thus heavily depending on large-scale labeled data, which could be infeasible when labels are prohibitively expensive. In this paper, we propose a novel Contrast-Reconstruction Representation Learning network (CRRL) that simultaneously captures postures and motion dynamics for unsupervised skeleton-based action recognition. It consists of three parts: Sequence Reconstructor (SER), Contrastive Motion Learner (CML), and Information Fuser (INF). SER learns representation from skeleton coordinate sequence via reconstruction. However the learned representation tends to focus on trivial postural coordinates and be hesitant in motion learning. To enhance the learning of motions, CML performs contrastive learning between the representation learned from coordinate sequences and additional velocity sequences, respectively. Finally, in the INF module, we explore varied strategies to combine SER and CML, and propose to couple postures and motions via a knowledge-distillation based fusion strategy which transfers the motion learning from CML to SER. Experimental results on several benchmarks, i.e., NTU RGB+D 60/120, PKU-MMD, CMU, and NW-UCLA, demonstrate the promise of the our method by outperforming state-of-the-art approaches.

With the deepening of the power market reform on the retail side, it is of great significance to study the economic optimization of the microgrid cluster system. Aiming at the economics of the microgrid cluster, comprehensively considering the degradation cost of energy storage battery, the compensation cost of demand-side controllable loads dispatch, the electricity transaction cost between the microgrids, and the electricity transaction cost between the microgrid and the power distribution network of the microgrid cluster, we establish an optimal dispatch model for the microgrid cluster including wind turbines, photovoltaics, and energy storage batteries. At the same time, in view of the problem that the population diversity of the basic sparrow search algorithm decreases and it is easy to fall into local extremes in the later iterations of the basic sparrow search algorithm, a chaos sparrow search algorithm based on Bernoulli chaotic mapping, dynamic adaptive weighting, Cauchy mutation, and reverse learning is proposed, and different types of test functions are used to analyze the convergence effect of the algorithm, and the calculation effects of the sparrow algorithm, the particle swarm algorithm, the chaotic particle swarm, and the genetic algorithm are compared. The algorithm has higher convergence speed, higher accuracy, and better global optimization ability. Finally, through the calculation example, it is concluded that the benefit of the microgrid cluster is increased by nearly 20%, which verifies the effectiveness of the improvement., Competing Interests: The authors declare that they have no conflicts of interest., (Copyright © 2021 Peng Wang et al.)

This paper presents a variant of multiscale quantum harmonic oscillator algorithm for multimodal optimization named MQHOA-MMO. MQHOA-MMO has only two main iterative processes: quantum harmonic oscillator process and multiscale process. In the two iterations, MQHOA-MMO only does one thing: sampling according to the wave function at different scales. A set of benchmark test functions including some challenging functions are used to test the performance of MQHOA-MMO. Experimental results demonstrate good performance of MQHOA-MMO in solving multimodal function optimization problems. For the 12 test functions, all of the global peaks can be found without being trapped in a local optimum, and MQHOA-MMO converges within 10 iterations.

Outbreaks of hand-foot-mouth disease (HFMD) have occurred many times and caused serious health burden in China since 2008. Application of modern information technology to prediction and early response can be helpful for efficient HFMD prevention and control. A seasonal auto-regressive integrated moving average (ARIMA) model for time series analysis was designed in this study. Eighty-four-month (from January 2009 to December 2015) retrospective data obtained from the Chinese Information System for Disease Prevention and Control were subjected to ARIMA modeling. The coefficient of determination (R 2 ), normalized Bayesian Information Criterion (BIC) and Q-test P value were used to evaluate the goodness-of-fit of constructed models. Subsequently, the best-fitted ARIMA model was applied to predict the expected incidence of HFMD from January 2016 to December 2016. The best-fitted seasonal ARIMA model was identified as (1,0,1)(0,1,1) 12 , with the largest coefficient of determination (R 2 =0.743) and lowest normalized BIC (BIC=3.645) value. The residuals of the model also showed non-significant autocorrelations (P Box-Ljung (Q) =0.299). The predictions by the optimum ARIMA model adequately captured the pattern in the data and exhibited two peaks of activity over the forecast interval, including a major peak during April to June, and again a light peak for September to November. The ARIMA model proposed in this study can forecast HFMD incidence trend effectively, which could provide useful support for future HFMD prevention and control in the study area. Besides, further observations should be added continually into the modeling data set, and parameters of the models should be adjusted accordingly.

We report a self-reference algorithm to discriminate human and nonhuman blood by calculating the ratios of identification Raman peaks to reference Raman peaks and choosing appropriate threshold values. The influence of using different reference peaks and identification peaks was analyzed in detail. The Raman peak at 1003  cm-1 was proved to be a stable reference peak to avoid the influencing factors, such as the incident laser intensity and the amount of sample. The Raman peak at 1341  cm-1 was found to be an efficient identification peak, which indicates that the difference between human and nonhuman blood results from the C-H bend in tryptophan. The comparison between self-reference algorithm and partial least square method was made. It was found that the self-reference algorithm not only obtained the discrimination results with the same accuracy, but also provided information on the difference of chemical composition. In addition, the performance of self-reference algorithm whose true positive rate is 100% is significant for customs inspection to avoid genetic disclosure and forensic science., ((2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE).)

It has become clear that repetitive sequences have played multiple roles in eukaryotic genome evolution including increasing genetic diversity through mutation, changes in gene expression and facilitating generation of novel genes. However, identification of repetitive elements can be difficult in the ab initio manner. Currently, some classical ab initio tools of finding repeats have already presented and compared. The completeness and accuracy of detecting repeats of them are little pool. To this end, we proposed a new ab initio repeat finding tool, named HashRepeatFinder, which is based on hash index and word counting. Furthermore, we assessed the performances of HashRepeatFinder with other two famous tools, such as RepeatScout and Repeatfinder, in human genome data hg19. The results indicated the following three conclusions: (1) The completeness of HashRepeatFinder is the best one among these three compared tools in almost all chromosomes, especially in chr9 (8 times of RepeatScout, 10 times of Repeatfinder); (2) in terms of detecting large repeats, HashRepeatFinder also performed best in all chromosomes, especially in chr3 (24 times of RepeatScout and 250 times of Repeatfinder) and chr19 (12 times of RepeatScout and 60 times of Repeatfinder); (3) in terms of accuracy, HashRepeatFinder can merge the abundant repeats with high accuracy.

A shortest path-based algorithm is proposed in this paper to find splitting lines for touching cell nuclei. First, an initial splitting line is obtained through the distance transform of a marker image and the watershed algorithm. The initial splitting line is then separated into different line segments as necessary, and the endpoint positions of these line segments are adjusted to the concave points on the contour. Finally, a shortest path algorithm is used to find the accurate splitting line between the starting-point and the end-point, and the final split can be achieved by the contour of the touching cell nuclei and the splitting lines. Comparisons of experimental results show that the proposed algorithm is effective for segmentation of different types of touching cell nuclei., (Copyright © 2014 Elsevier Ltd. All rights reserved.)

Pose estimation aims at measuring the position and orientation of a calibrated camera using known image features. The pinhole model is the dominant camera model in this field. However, the imaging precision of this model is not accurate enough for an advanced pose estimation algorithm. In this paper, a new camera model, called incident ray tracking model, is introduced. More importantly, an advanced pose estimation algorithm based on the perspective ray in the new camera model, is proposed. The perspective ray, determined by two positioning points, is an abstract mathematical equivalent of the incident ray. In the proposed pose estimation algorithm, called perspective-ray-based scaled orthographic projection with iteration (PRSOI), an approximate ray-based projection is calculated by a linear system and refined by iteration. Experiments on the PRSOI have been conducted, and the results demonstrate that it is of high accuracy in the six degrees of freedom (DOF) motion. And it outperforms three other state-of-the-art algorithms in terms of accuracy during the contrast experiment.

A famous biologically inspired hierarchical model (HMAX model), which was proposed recently and corresponds to V1 to V4 of the ventral pathway in primate visual cortex, has been successfully applied to multiple visual recognition tasks. The model is able to achieve a set of position- and scale-tolerant recognition, which is a central problem in pattern recognition. In this paper, based on some other biological experimental evidence, we introduce the memory and association mechanism into the HMAX model. The main contributions of the work are: 1) mimicking the active memory and association mechanism and adding the top down adjustment to the HMAX model, which is the first try to add the active adjustment to this famous model and 2) from the perspective of information, algorithms based on the new model can reduce the computation storage and have a good recognition performance. The new model is also applied to object recognition processes. The primary experimental results show that our method is efficient with a much lower memory requirement.

Metamodels have been widely used in engineering design to facilitate analysis and optimization of complex systems that involve computationally expensive simulation programs. The accuracy of metamodels is directly related to the experimental designs used. Optimal Latin hypercube designs are frequently used and have been shown to have good space-filling and projective properties. However, the high cost in constructing them limits their use. In this paper, a methodology for creating novel Latin hypercube designs via translational propagation and successive local enumeration algorithm (TPSLE) is developed without using formal optimization. TPSLE algorithm is based on the inspiration that a near optimal Latin Hypercube design can be constructed by a simple initial block with a few points generated by algorithm SLE as a building block. In fact, TPSLE algorithm offers a balanced trade-off between the efficiency and sampling performance. The proposed algorithm is compared to two existing algorithms and is found to be much more efficient in terms of the computation time and has acceptable space-filling and projective properties.

Analysis of vessel structures in 2D X-ray angiograms is important for pre-operative evaluation and image-guided intervention. However, automated vessel segmentation in angiograms, especially extraction of the topology such as bifurcations and vessel crossings, remains challenging mainly due to the projective nature of angiography and background clutter. In this paper, a novel framework for model-guided coronary vessel extraction in 2D angiograms is presented. In this framework, a graph is constructed using a sparse set of pixels in the angiogram. With a single user-supplied click as the starting point, the vessel tree structure in the angiogram is automatically extracted from the graph. Ambiguities in this tree structure caused by 3D-to-2D projection are then resolved using topological information from the 3D vessel model of the same patient. By incorporating this prior shape information, the proposed method is effective in extraction of vessel topology, and is robust to background clutter and uneven illumination. Through quantitative evaluation on 20 angiograms, it is shown that this model-guided approach significantly improves detection of vessel structures and bifurcations.

This paper presents a novel biologically inspired metaheuristic algorithm called seven-spot ladybird optimization (SLO). The SLO is inspired by recent discoveries on the foraging behavior of a seven-spot ladybird. In this paper, the performance of the SLO is compared with that of the genetic algorithm, particle swarm optimization, and artificial bee colony algorithms by using five numerical benchmark functions with multimodality. The results show that SLO has the ability to find the best solution with a comparatively small population size and is suitable for solving optimization problems with lower dimensions.

The purpose of this study was to develop a technique to measure the dielectric properties of biological tissues with an interstitial dipole antenna based upon previous efforts for open-ended coaxial probes. The primary motivation for this technique is to facilitate treatment monitoring during microwave tumor ablation by utilizing the heating antenna without additional intervention or interruption of the treatment. The complex permittivity of a tissue volume surrounding the antenna was calculated from reflection coefficients measured after high-temperature microwave heating by using a rational function model of the antenna's input admittance. Three referencing liquids were needed for measurement calibration. The dielectric measurement technique was validated ex vivo in normal and ablated bovine livers. Relative permittivity and effective conductivity were lower in the ablation zone when compared to normal tissue, consistent with previous results. The dipole technique demonstrated a mean 10% difference of permittivity values when compared to open-ended coaxial cable measurements in the frequency range of 0.5-20 GHz. Variability in measured permittivities could be smoothed by fitting to a Cole-Cole dispersion model. Further development of this technique may facilitate real-time monitoring of microwave ablation treatments through the treatment applicator., (© 2011 IEEE)

Successful predictions of peptide MHC binding typically require a large set of binding data for the specific MHC molecule that is examined. Structure based prediction methods promise to circumvent this requirement by evaluating the physical contacts a peptide can make with an MHC molecule based on the highly conserved 3D structure of peptide:MHC complexes. While several such methods have been described before, most are not publicly available and have not been independently tested for their performance. We here implemented and evaluated three prediction methods for MHC class II molecules: statistical potentials derived from the analysis of known protein structures; energetic evaluation of different peptide snapshots in a molecular dynamics simulation; and direct analysis of contacts made in known 3D structures of peptide:MHC complexes. These methods are ab initio in that they require structural data of the MHC molecule examined, but no specific peptide:MHC binding data. Moreover, these methods retain the ability to make predictions in a sufficiently short time scale to be useful in a real world application, such as screening a whole proteome for candidate binding peptides. A rigorous evaluation of each methods prediction performance showed that these are significantly better than random, but still substantially lower than the best performing sequence based class II prediction methods available. While the approaches presented here were developed independently, we have chosen to present our results together in order to support the notion that generating structure based predictions of peptide:MHC binding without using binding data is unlikely to give satisfactory results.

An accurate and robust method to detect curve structures, such as a vessel branch or a guidewire, is essential for many medical imaging applications. A fully automatic method, although highly desired, is prone to detection errors that are caused by image noise and curve-like artifacts. In this paper, we present a novel method to interactively detect a curve structure in a 2D fluoroscopy image with a minimum requirement of human corrections. In this work, a learning based method is used to detect curve segments. Based on the detected segment candidates, a graph is built to search a curve structure as the best path passing through user interactions. Furthermore, our method introduces a novel hyper-graph based optimization method to allow for imposing geometric constraints during the path searching, and to provide a smooth and quickly converged result. With minimum human interactions involved, the method can provide accurate detection results, and has been used in different applications for guidewire and vessel detections.

Purpose: To demonstrate the efficacy of an automated three-dimensional (3D) template matching-based algorithm in detecting brain metastases on conventional MR scans and the potential of our algorithm to be developed into a computer-aided detection tool that will allow radiologists to maintain a high level of detection sensitivity while reducing image reading time., Materials and Methods: Spherical tumor appearance models were created to match the expected geometry of brain metastases while accounting for partial volume effects and offsets due to the cut of MRI sampling planes. A 3D normalized cross-correlation coefficient was calculated between the brain volume and spherical templates of varying radii using a fast frequency domain algorithm to identify likely positions of brain metastases., Results: Algorithm parameters were optimized on training datasets, and then data were collected on 22 patient datasets containing 79 total brain metastases producing a sensitivity of 89.9% with a false positive rate of 0.22 per image slice when restricted to the brain mass., Conclusion: Study results demonstrate that the 3D template matching-based method can be an effective, fast, and accurate approach that could serve as a useful tool for assisting radiologists in providing earlier and more definitive diagnoses of metastases within the brain., ((c) 2009 Wiley-Liss, Inc.)

This paper presents a new technique of coronary digital subtraction angiography which separates layers of moving background structures from dynamic fluoroscopic sequences of the heart and obtains moving layers of coronary arteries. A Bayeisan framework combines dense motion estimation, uncertainty propagation and statistical fusion to achieve reliable background layer estimation and motion compensation for coronary sequences. Encouraging results have been achieved on clinically acquired coronary sequences, where the proposed method considerably improves the visibility and perceptibility of coronary arteries undergoing breathing and cardiac movements. Perceptibility improvement is significant especially for very thin vessels. Clinical benefit is expected in the context of obese patients and deep angulation, as well as in the reduction of contrast dose in normal size patients.

Significant appearance changes of objects under different orientations could cause loss of tracking, "drifting." In this paper, we present a collaborative tracking framework to robustly track faces under large pose and expression changes and to learn their appearance models online. The collaborative tracking framework probabilistically combines measurements from an offline-trained generic face model with measurements from online-learned specific face appearance models in a dynamic Bayesian network. In this framework, generic face models provide the knowledge of the whole face class, while specific face models provide information on individual faces being tracked. Their combination, therefore, provides robust measurements for multiview face tracking. We introduce a mixture of probabilistic principal component analysis (MPPCA) model to represent the appearance of a specific face under multiple views, and we also present an online EM algorithm to incrementally update the MPPCA model using tracking results. Experimental results demonstrate that the collaborative tracking and online learning methods can handle large pose changes and are robust to distractions from the background.

The identification of MHC class II restricted peptide epitopes is an important goal in immunological research. A number of computational tools have been developed for this purpose, but there is a lack of large-scale systematic evaluation of their performance. Herein, we used a comprehensive dataset consisting of more than 10,000 previously unpublished MHC-peptide binding affinities, 29 peptide/MHC crystal structures, and 664 peptides experimentally tested for CD4+ T cell responses to systematically evaluate the performances of publicly available MHC class II binding prediction tools. While in selected instances the best tools were associated with AUC values up to 0.86, in general, class II predictions did not perform as well as historically noted for class I predictions. It appears that the ability of MHC class II molecules to bind variable length peptides, which requires the correct assignment of peptide binding cores, is a critical factor limiting the performance of existing prediction tools. To improve performance, we implemented a consensus prediction approach that combines methods with top performances. We show that this consensus approach achieved best overall performance. Finally, we make the large datasets used publicly available as a benchmark to facilitate further development of MHC class II binding peptide prediction methods.

This paper presents methods of modeling and predicting face recognition (FR) system performance based on analysis of similarity scores. We define the performance of an FR system as its recognition accuracy, and consider the intrinsic and extrinsic factors affecting its performance. The intrinsic factors of an FR system include the gallery images, the FR algorithm, and the tuning parameters. The extrinsic factors include mainly query image conditions. For performance modeling, we propose the concept of "perfect recognition," based on which a performance metric is extracted from perfect recognition similarity scores (PRSS) to relate the performance of an FR system to its intrinsic factors. The PRSS performance metric allows tuning FR algorithm parameters offline for near optimal performance. In addition, the performance metric extracted from query images is used to adjust face alignment parameters online for improved performance. For online prediction of the performance of an FR system on query images, features are extracted from the actual recognition similarity scores and their corresponding PRSS. Using such features, we can predict online if an individual query image can be correctly matched by the FR system, based on which we can reduce the incorrect match rates. Experimental results demonstrate that the performance of an FR system can be significantly improved using the presented methods.

Independent component analysis (ICA) is a new method of signal statistical processing and widely used in many fields. We face several problems such as the different nature of source signals (e.g. both super-Gaussian and sub-Gaussian sources exist), unknown number of sources and contamination of the sensor signals with a high level of additive noise in the analysis of signal. A robust approach was proposed to solve these problems in this paper. Firstly, observations (noisy data) possessing high dimensionality were preprocessed and decomposed into a source signal subspace and a noise subspace. Then the number of sources was got through the cross-validation method, and this solved the problem that ICA could not confirm the number of sources. At last the transformed low-dimensional source signals were further separated with the fast and stable ICA algorithm. Through the analysis of artificially synthesized data and the real-world Magnetoencephalographic data, the efficacy of this robust approach was illustrated.

In this paper, the discrete octonion linear canonical transform (DOCLCT) is presented. According to the definition of DOCLCT, its shift transform and modulation transform are explored. In addition, based on a new convolution operator, we obtain the convolution theorem of DOCLCT. The DOCLCT can be derived by the three-dimensional (3D) linear canonical transform. So, the fast algorithm of 3D DOCLCT is discussed. Finally, Heisenberg's uncertainty principle associated with DOCLCT is established. [ABSTRACT FROM AUTHOR]

Subpixel water area mapping based on spatial information processing at superpixel scale (SISS) for multispectral imagery is proposed in this paper. In SISS, the superpixels with irregular shape are first obtained by segmenting the first principal component of the improved imagery which is improved from the the original coarse multispectral imagery by using bicubic interpolation; The random walk algorithm is then introduced to calculate these superpixels to obtain the spatial information. Finally, according to this spatial information, the label of the water area or the non-water area is assigned to each subpixel by the class allocation method, producing the final mapping result. The experimental results on two datasets show that the proposed method can achieve the best performance in both visual effects and quantitative indicators. [ABSTRACT FROM AUTHOR]

The method of quantum dynamics is employed to investigate the mean strategy in the swarm intelligence algorithm. The physical significance of the population mean point is explained as the location where the optimal solution with the highest likelihood can be found once a quantum system has reached a ground state. Through the use of the double well function and the CEC2013 test suite, controlled experiments are conducted to perform a comprehensive performance analysis of the mean strategy. The empirical results indicate that implementing the mean strategy not only enhances solution diversity but also yields accurate, efficient, stable, and effective outcomes for finding the optimal solution. [ABSTRACT FROM AUTHOR]

Terrain visibility analysis is vital in geospatial research, employing computer geometry and graphics principles for computing and visualizing visibility between observation and target points. The observation point setup problem is crucial for tasks like sentry point selection and signal base station placement. This problem involves choosing the minimum viewpoints on terrain for optimal joint view coverage, presenting a combinatorial optimization challenge. As technology advances, data can be acquired with increasing precision, the number of viewpoints that can be extracted from the same area is gradually increasing as well, obtaining candidate viewpoints and determining optimal combinations become challenging. This paper proposes a novel method, Stepwise Maximum Viewshed (SMV), addressing observation point setup by stepwise filtering the maximum viewshed and can customize the size of the observation point combination. In complex mountainous terrain, the SMV algorithm demonstrates superior joint view coverage compared to Candidate Viewpoints Filtering (CVF) and Simulated Annealing (SA) algorithms. Experimental results reveal up to 5.59% improvement over CVF and a maximum of 12.52% over SA in joint viewshed coverage. [ABSTRACT FROM AUTHOR]

Due to the complex underwater environment and the selective absorption and scattering effect of water on light waves, underwater images often suffer from issues such as low contrast, color distortion, and blurred details. This paper presents a stable and effective algorithm for enhancing underwater images to address these challenges. Firstly, an improved color correction algorithm based on the gray world and minimum information loss is employed to remove the blue-green bias present in the images. Secondly, a contrast enhancement algorithm is based on the guided filter and wavelet decomposition to make the texture details of the image clearer. Then, the normalized weight map of the image is obtained to carry out multi-scale fusion. Finally, the fused image is applied to perform the multi-scale decomposition. The experimental results show that the algorithm proposed in this paper can correct the image color deviation, improve the image contrast and enhance the image details. [ABSTRACT FROM AUTHOR]

YOLOv7 is one of the most effective algorithms for one‐stage detectors. However, when it is applied to pedestrian detection tasks in the industrial scene, it is still challenging for complex environments and multi‐scale changes of pedestrians. This paper proposes a new pedestrian detector for the industrial scene based on improved YOLOv7‐tiny and named as GP‐YOLO. First, the neck of YOLOv7‐tiny is replaced by RepGFPN structure, make full use of multi‐scale features to enhance the detection accuracy of objects with large‐scale changes. Second, a new gnconv branch is added to the feature fusion module, and the high‐order spatial interaction capability is introduced to further enhance the target detection accuracy. Finally, a lightweight method based on PModule is proposed, on this basis, a PConv bottleneck is designed to reduce the FLOPs and enhance the feature extraction. Experiments on a self‐made Industrial Pedestrian Data set show that before lightweight, the proposed algorithm achieves a 3.2% improvement in mAP@0.5:0.95 and a 3.7% improvement in Recall compared to the baseline YOLOv7‐tiny. After lightweight GP‐YOLO, compared to non‐lightweight, parameters and FLOPs are decreased by 26% and 23%, respectively, the mAP@0.5:0.95 is decreased by only 1.1% and the Recall is decreased by only 1.3%, which remains at a high level. Compared with baseline YOLOv7‐tiny, the lightweight GP‐YOLO has similar parameters and FLOPs, but the mAP@0.5:0.95 is increased by 2.1%, and the Recall is increased by 2.4%. © 2024 Institute of Electrical Engineers of Japan and Wiley Periodicals LLC. [ABSTRACT FROM AUTHOR]

In this paper, a random derivative-free modified tensor method with curvilinear linesearch technique is considered for solving nonlinear programming problems. The proposed algorithm is designed to build polynomial interpolation models for the objective function and build the tensor model using the information of the interpolation function. At the same time, we give a new curvilinear tensor step which guarantees the monotonic decrease on the tensor model. The modified tensor step also asymptotically approaches the modified Newton direction as the step length shrinks to zero, and the objective function of problem will be descendent. Under general assumptions, we give the global and local superlinear convergence of the algorithm. Numerical results are recorded, and the compare results with a tensor algorithm without curvilinear linesearch technique and Newton algorithm show that our algorithm is more effectiveness. [ABSTRACT FROM AUTHOR]

Background: The burden of lumbar degenerative diseases (LDD) has increased substantially with the unprecedented aging population. Identifying elderly patients with high risk of postoperative adverse events (AEs) and establishing individualized perioperative management is critical to mitigate added costs and optimize cost-effectiveness to the healthcare system. We aimed to develop a predictive tool for AEs in elderly patients with transforaminal lumbar interbody fusion (TLIF), utilizing multivariate logistic regression, single classification and regression tree (hereafter, "classification tree"), and random forest machine learning algorithms. Methods: This study was a retrospective review of a prospective Geriatric Lumbar Disease Database (age ≥ 65). Our outcome measure was postoperative AEs, including prolonged hospital stays, postoperative complications, readmission, and reoperation within 90 days. Patients were grouped as either having at least one adverse event (AEs group) or not (No-AEs group). Three models for predicting postoperative AEs were developed using training dataset and internal validation using testing dataset. Finally, online tool was developed to assess its validity in the clinical setting (external validation). Results: The development set included 1025 patients (mean [SD] age, 72.8 [5.6] years; 632 [61.7%] female), and the external validation set included 175 patients (73.2 [5.9] years; 97 [55.4%] female). The predictive ability of our three models was comparable, with no significant differences in AUC (0.73 vs. 0.72 vs. 0.70, respectively). The logistic regression model had a higher net benefit for clinical intervention than the other models. A nomogram based on logistic regression was developed, and the C-index of external validation for AEs was 0.69 (95% CI 0.65–0.76). Conclusion: The predictive ability of our three models was comparable. Logistic regression model had a higher net benefit for clinical intervention than the other models. Our nomogram and online tool (https://xuanwumodel.shinyapps.io/Model%5ffor%5fAEs/) could inform physicians about elderly patients with a high risk of AEs within the 90 days after TLIF surgery. [ABSTRACT FROM AUTHOR]

The pursuit of higher recognition accuracy and speed with smaller model sizes has been a major research topic in the detection of surface defects in steel. In this paper, we propose an improved high-speed and high-precision Efficient Fusion Coordination network (EFC-YOLO) without increasing the model's size. Since modifications to enhance feature extraction in shallow networks tend to affect the speed of model inference, in order to simultaneously ensure the accuracy and speed of detection, we add the improved Fusion-Faster module to the backbone network of YOLOv7. Partial Convolution (PConv) serves as the basic operator of the module, which strengthens the feature-extraction ability of shallow networks while maintaining speed. Additionally, we incorporate the Shortcut Coordinate Attention (SCA) mechanism to better capture the location information dependency, considering both lightweight design and accuracy. The de-weighted Bi-directional Feature Pyramid Network (BiFPN) structure used in the neck part of the network improves the original Path Aggregation Network (PANet)-like structure by adding step branches and reducing computations, achieving better feature fusion. In the experiments conducted on the NEU-DET dataset, the final model achieved an 85.9% mAP and decreased the GFLOPs by 60%, effectively balancing the model's size with the accuracy and speed of detection. [ABSTRACT FROM AUTHOR]

A novel breast ultrasound tomography system based on a circular array of capacitive micromechanical ultrasound transducers (CMUT) has broad application prospects. However, the images produced by this system are not suitable as input for the training phase of the super-resolution (SR) reconstruction algorithm. To solve the problem, this paper proposes an improved medical image super-resolution (MeSR) method based on the sparse domain. First, we use the simultaneous algebraic reconstruction technique (SART) with high imaging accuracy to reconstruct the image into a training image in a sparse domain model. Secondly, we denoise and enhance the contrast of the SART images to obtain improved detail images before training the dictionary. Then, we use the original detail image as the guide image to further process the improved detail image. Therefore, a high-precision dictionary was obtained during the testing phase and applied to filtered back projection SR reconstruction. We compared the proposed algorithm with previously reported algorithms in the Shepp Logan model and the model based on the CMUT background. The results showed significant improvements in peak signal-to-noise ratio, entropy, and average gradient compared to previously reported algorithms. The experimental results demonstrated that the proposed MeSR method can use noisy reconstructed images as input for the training phase of the SR algorithm and produce excellent visual effects. [ABSTRACT FROM AUTHOR]

Multi-terminal DC transmission (MTDC) systems have attracted much attention due to their significant advantages in long-distance and high-capacity transmission. To improve their reliability and operation performance, a distributed fixed-time secondary control of frequency restoration and active power sharing is proposed under event-triggered communication, which only depends on the states of each AC grid and its neighbors. By utilizing Lyapunov theory, we prove that the MTDC system with the fixed-time secondary control can be stable in a settling time, and the conditions of the settling time are established for fixed-time algorithms. In addition, we simulate a five-terminal MTDC system in Matlab/Simulink. Several cases of MTDC systems are exhibited to showcase how well the suggested controller works when dealing with load changes and attacks. The comparison of the number of event-triggered instants shows that the proposed control method can effectively reduce communication resources. [ABSTRACT FROM AUTHOR]

In the prediction and modeling analysis of wear degree in the field of industrial parts processing, there are problems such as poor prediction ability for long sequence data and low sensitivity of output feedback to changes in input signals. In this paper, a combined prediction model is proposed that integrates dual attention mechanisms and self-regressive correction. Firstly, pre-processing is performed on the collected wear data to eliminate noise and aberrant mutation data. Then, the feature attention mechanism is introduced to analyze the input data sequence, and the weights of each feature under the temporal condition are set based on the contribution of the prediction results, thereby obtaining the LSTM hidden state at the current time. Subsequently, the temporal attention mechanism is introduced to perform a weighted calculation of the hidden state information, analyze the correlation of long-term sequential wear data, and decode and output the analysis results. Finally, the ARIMA model is used to perform linear correction on the predicted results to improve the accuracy of wear degree prediction. The proposed model is compared and analyzed with the models that are highly related in recent research on real-world wear degree datasets. The experimental results show that the improved model has a better ability to improve the corresponding problems and has a significant increase in prediction accuracy. [ABSTRACT FROM AUTHOR]

An autonomous optimal trajectory planning method based on the deep deterministic policy gradient (DDPG) algorithm of reinforcement learning (RL) for hypersonic vehicles (HV) is proposed in this paper. First, the trajectory planning problem is converted into a Markov Decision Process (MDP), and the amplitude of the bank angle is designated as the control input. The reward function of the MDP is set to minimize the trajectory terminal position errors with satisfying hard constraints. The deep neural network (DNN) is used to approximate the policy function and action-value function in the DDPG framework. The Actor network then computes the control input directly according to flight states. Using a limited exploration strategy, the optimal policy network would be considered fully trained when the reward value reached maximum convergence. Simulation results show that the policy network trained using a DDPG algorithm accomplishes 3-dimensional (3D) trajectory planning during the HV glide phase with high terminal precision and stable convergence. Additionally, the single step calculation time of the policy network occurs in near real time, which suggests great potential as an autonomous online trajectory planner. Monte Carlo experiments prove the strong robustness of the implementation of an autonomous trajectory planner under aerodynamic disturbances. [ABSTRACT FROM AUTHOR]

Due to the high cost of annotating dense fruit images, annotated target images are limited in some ripeness detection applications, which significantly restricts the generalization ability of small object detection networks in complex environments. To address this issue, this study proposes a self-supervised cherry ripeness detection algorithm based on multi-feature contrastive learning, consisting of a multi-feature contrastive self-supervised module and an object detection module. The self-supervised module enhances features of unlabeled fruit images through random contrastive augmentation, reducing interference from complex backgrounds. The object detection module establishes a connection with the self-supervised module and designs a shallow feature fusion network based on the input target scale to improve the detection performance of small-sample fruits. Finally, extensive experiments were conducted on a self-made cherry dataset. The proposed algorithm showed improved generalization ability compared to supervised baseline algorithms, with better accuracy in terms of mAP, particularly in detecting distant small cherries. [ABSTRACT FROM AUTHOR]

X‐ray flaw detection is widely used in non‐destructive area. The intuitive defect information can be obtained through the X‐ray film, which is usually digitized into high greyscale image by a 12‐bit or 16‐bit, called super 8‐bit, industrial scanner. When an ordinary 8‐bit monitor displays a super 8‐bit greyscale image, it appears loss of detail information, blurring of the image appears and other problems. Therefore, in this paper, a pseudo‐colour enhancement algorithm for displaying high‐bit RAW images on low‐bit monitor was proposed according to the chromatographic mapping relationship based on the visual characteristics of human eye. First, a high grey‐scale pseudo‐colour enhancement algorithm, called HOTM‐HGL, based on hot metal coding was proposed based on the standard film, whose enhancement effect is better than current mainstream algorithms. Second, aiming at the non‐standard film, a new stretching function called RAW‐Optical‐Stretching was reconstructed to improve HOTM‐HGL algorithm, called HOTM‐HGLS algorithm, whose display effect on ordinary monitors was improved in further. Finally, HOTM‐HGLS algorithm was applied in the detection of X‐ray film defect, which was convenient for the capture of defect information in the weld. Compared with the existing algorithms, various indicators have been greatly improved, enriching the amount of information and strengthening the image recognition effects. [ABSTRACT FROM AUTHOR]