Search

Spectral and texture features play important roles in plantation leaf area index (LAI) estimation, and their combination may enhance LAI inversion accuracy. Furthermore, research on the impact of different machine learning (ML) models on their hyperparameter combinations and splitting ratios remains challenging. In our study, experiments based on spectral and textural features of GF-6 WFV data were conducted on Eucalyptus grandis plantation forests in Huangmian Town, Guangxi, China. ML methods such as multiple stepwise regression (MSR), random forest (RF), back-propagation neural network (BPNN), and support vector regression (SVR) were mainly utilized to perform model hyper-parameter tuning and split-ratio analysis in order to estimate the LAI. The results of the study showed that spectral and gray level co-occurrence matrix (GLCM) texture features were very sensitive to changes in Eucalyptus grandis LAI. The accuracy of combining the two was 10% higher than when they were not combined. Furthermore, it was found that the nonlinear methods (RF, BPNN, and SVR) outperformed the linear method (MSR), with the average 2 Rmax of the nonlinear model being 26% higher than that of the linear model, and the RMSE value being 29% lower than that of the linear model. In addition, by analyzing different combinations of features, model hyperparameter fine-tuning, and splitting ratios in the nonlinear model, it was found that the splitting ratios of different combinations of model hyperparameters have a great impact on the accuracy of the model. A total of 12 out of 21 data sets showed high accuracy and stability at a split ratio of 8.5:1.5 (ratio of 0.85), with the best-performing RF model differing from the lowest by 91% for 2 max R and 39% for 2 Rstd. Combining spectral and texture features provides highly accurate inversion data. Model hyper-parameter fine-tuning and segmental scale tuning can facilitate the application of inversion data to fully utilize the optimal performance of the ML model. [ABSTRACT FROM AUTHOR]

The timely detection and identification of crop diseases is a crucial aspect of the agricultural sector. It contributes significantly to the by and large productivity of the plant. One of the most crucial factors that we need to consider while determining a plant's susceptibility to a particular disease is the visual characteristics of the affected plant. The increasing popularity of automation and availability of efficient techniques for disease identification has led to the development of novel methods and engraved impactful technologies in field of automated disease detection. The traditional methods have not been able to provide the researchers with the most accurate results. The proposed model in this work can identify the rice crop disease without relying on subjective data and have many advantages over traditional approaches as evident from the results derived. It has the potential to improve the efficiency of the process and aid in early detection. Machine learning method presents real-time automated decision support systems and can help improve crop or plant growth productivity and quality. This work aims to introduce a new and enhanced method as Neuro-GA, which is a combination of both the artificial neural network (ANN) and the genetic algorithm (GA). It has been claimed that it is more powerful and accurate than the traditional methods. The pioneer and nascent stages of this analysis includes preprocessing of the data was carried out. The features were then extracted using Gray-level co-occurrence matrix (GLCM) and subsequently the finally extracted features were cascaded to the Neuro-GA classifier. The digital image processing (DIP) techniques used in this study for rendering visual images along with Neuro-GA classifier resulted in skyrocket accuracy level of 90% and above. The technique validated in this study has allowed the automated monitoring of various aspects of crop production and farming and an omnipotent promising efficiency hence this approach can be magnanimously effective in monitoring agricultural production and thereby plummeting waste allied with crop damage. [ABSTRACT FROM AUTHOR]

Lung cancer is relying as one of the significant and leading cause for the deaths which are based on cancer. So, an effective diagnosis is a crucial step to save the patients who are all dying due to lung cancer. Moreover, the diagnosis must be performed based on the severity of lung cancer and the severity can be addressed with the help of an optimal classification approach. So, this research introduced an Attention based Capsule Network (A‐Caps Net) with dispersed dynamic routing to perform in‐depth classification of the disease affected partitions of the image and results in better classification results. The attention layer with dispersed dynamic routing evaluates the digit capsule from feature vector in a constant manner. As the first stage, data acquisitioned from datasets such as Lung Nodule Analysis‐16 (LUNA‐16), The Cancer Imaging Archive (TCIA) dataset and Lung Image Database Consortium and Image Database Resource Initiative (LIDC‐IDRI). After acquisitioning data, pre‐processing is done to enhance the resolution of the image using Generative Adversarial Network. The pre‐processed output is given as output for extraction of features that takes place using GLCM and VGG‐16 which extracts the low level features and high level features respectively. Finally, categorization of lung cancer is performed using Attention based Capsule Network (A‐Caps Net) with dispersed dynamic routing which categorize the lung cancer as benign and malignant. The results obtained through experimental analysis exhibits that proposed approach attained better accuracy of 99.57%, 99.91% and 99.29% for LUNA‐16, LIDC‐IDRI and TCIA dataset respectively. The classification accuracy achieved by the proposed approach for LUNA‐16 dataset is 99.57% which is comparably higher than DBN, 3D CNN, Squeeze Nodule Net and 3D‐DCNN with multi‐layered filter with accuracies of 99.16%, 97.17% and 94.1% respectively. [ABSTRACT FROM AUTHOR]

Yao Hu,1,* Jiao Ai2,* 1Department of Obstetrics and Gynecology, Jingzhou Hospital Affiliated to Yangtze University,Jingzhou Central Hospital, Jingzhou, Hubei, People’s Republic of China; 2Department of Urology, Jingzhou Hospital Affiliated to Yangtze University, Jingzhou Central Hospital, Jingzhou, Hubei, People’s Republic of China*These authors contributed equally to this workCorrespondence: Yao Hu, Email huyao3784609@163.comObjective: To develop an early warning system that enables accurate parametrial invasion (PMI) risk prediction in cervical cancer patients with early-stage.Methods: We retrospectively collected 218 early-stage cervical cancer patients who were treated in Jingzhou Central Hospital from January 31, 2015, to January 31, 2023, and diagnosed with early stage cervical cancer by pathology. The prediction model training is achieved by randomly dividing 70% of the training queue population, with the remaining 30% used as the testing queue. Then, a prediction model based on machine learning algorithms (including random forest, generalized linear regression, decision tree, support vector machine, and artificial neural network) is constructed to predict the risk of PMI occurrence. Ultimately, the analysis of receiver operating characteristic curve (ROC) and decision curve analysis (DCA) is used to evaluate the predictive ability of various prediction models.Results: We finally included radiomics-based candidate variables that can be used for PMI model. Multivariate logistic regression analysis showed that energy, correlation, sum entropy (SUE), entropy, mean sum (MES), variance of differences (DIV), and inverse difference (IND) were independent risk factors for PMI occurrence. The predictive performance AUC of five types of machine learning ranges from 0.747 to 0.895 in the training set and can also reach a high accuracy of 0.905 in the testing set, indicating that the predictive model has ideal robustness.Conclusion: Our ML-based model incorporating GLCM parameters can predict PMI in cervical cancer patients with stage IB1 to IIA2, particularly the RFM, which could contribute to distinguishing PMI before surgery, especially in assisting decision-making on surgical scope.Keywords: cervical cancer, machine learning, parametrial invasion, gray level co-occurrence matrix, prediction model

Blindness refers to a condition in which an individual experiences limitations in their visual ability. Individuals with visual impairments require specific assistance to facilitate their movement from one location to another. The need for this assistance arises due to various obstacles scattered throughout their environment. One of the most significant challenges is navigating descending stairs. To address this issue, a descending stairs detection system based on digital image processing has been developed. Through this approach, the mobility of individuals with visual impairments can be enhanced. The descending stairs detection system is designed using the Gray Level Co-occurrence Matrix (GLCM) method to extract distinctive features of descending stairs and the surrounding floor surfaces. Seven GLCM features are incorporated into the development of this system, allowing it to differentiate between descending stairs and floors using the Extra Tree Classifier classification method. Through a series of tests, the system's accuracy is measured at 84%, demonstrating its adequate ability to identify descending stairs. Additionally, the average computation time for detecting descending stairs and floors is recorded at 0.121 (s), indicating the efficient performance of this system in supporting the mobility of individuals in need.

Detecting descending stairs and floors is a crucial aspect of implementing autonomous systems in smart wheelchairs. When the obstacle detection system used in wheelchairs fails to accurately identify descending stairs, it can lead to severe consequences for users, including injuries or, in the worst-case scenario, fatal accidents. Therefore, there is a pressing need for an algorithm that not only exhibits high accuracy in detecting obstacles on descending stairs but also operates with minimal computational delay to ensure an immediate response in wheelchair braking. In this research, We utilize the GLCM technique to extract texture characteristics. Out of these methods, the Decision Tree exhibits the highest accuracy, reaching 94%, with a remarkably fast computational time of 0.01299 s. These promising results were achieved by utilizing the GLCM method with a distance of 2 and an angle of 45°. The accuracy obtained has increased by 2.5% compared to the previous research. Such a level of accuracy, coupled with fast computational performance, enables smart wheelchairs to effectively assist users in identifying obstacles while descending stairs.

This research designs a touchless fingerprint identification biometric system. This research problem stems from the user's need to change the conventional system to touchless in the hope of minimizing direct contact that can be scanned by someone who is not interested. This research aims to implement palm images in validating identity using GLCM (Gray-Level Co-occurrence Matrix) features with the K-Nearest Neighbours (K-NN) classification method in MATLAB. The identification process is divided into several stages: image acquisition, pre-processing, feature extraction with GLCM, and database matching using KNN classification. System testing uses the 10-fold cross validation method with 100 image samples (90 training images and 10 test images) that are tested in turn to calculate the average accuracy and analyze system performance. Furthermore, the test used GLCM angles (0o, 45o, 90o dan 135o) and K-NN with k values of 1, 5 and 7. The results showed the highest accuracy of 72% using an angle of 0° in GLCM and k=1 and the lowest at an angle of 90o and k=7 in K-NN. The advantage of this design is the recognition of the identity of the fingerprint owner in real time.

Abstract Due to the limited availability of natural resources, it is essential that agricultural productivity keep pace with population growth. Despite unfavorable weather circumstances, this project's major objective is to boost production. As a consequence of technological advancements in agriculture, precision farming as a way for enhancing crop yields is gaining appeal and becoming more prevalent. When it comes to predicting future data, machine learning employs a number of methods, including the creation of models and the acquisition of prediction rules based on past data. In this manuscript, we examine various techniques to machine learning, as well as an automated agricultural yield projection model based on selecting the most relevant features. For the purpose of selecting features, the Grey Level Co-occurrence Matrix method is utilised. For classification, we make use of the AdaBoost Decision Tree, Artificial Neural Network (ANN), and K-Nearest Neighbour (KNN) algorithms. The data set that was used in this study is simply a compilation of information about a variety of topics, including yield, pesticide use, rainfall, and average temperature. This data collection consists of 33 characteristics or qualities in total. The crops soya beans, maze, potato, rice, paddy, wheat, and sorghum are included in this data collection. This data collection was made possible through the collaboration of the Food and Agriculture Organisation (FAO) and the World Data Bank, both of which make their data available to the public. The AdaBoost decision tree has achieved the highest level of accuracy possible when used to anticipate agricultural yield. Both the accuracy rate and the recall rate are quite high at 99 percent.

Objective: To achieve surface wear life prediction of abrasive blast rollers of grinding machines. Methods: The wear images of the grinding roller surface were acquired by the built image acquisition system, and the texture parameters such as second order moments, entropy value, contrast and correlation in the wear cycle of the grinding roller were obtained based on the grey scale co-generation matrix algorithm, and the obtained texture feature parameters were input into the constructed PSO-based LS-SVM algorithm model to finally predict the wear life of the blast roller. Results: The particle swarm algorithm could optimize the penalty factor and kernel parameters of LS-SVM well, and the PSO-LS-SVM algorithm was far superior to the LS-SVM algorithm model. The wear state of the blast roller surface of the mill could be accurately identified using the PSO-LS-SVM algorithm. Conclusion: The system can accurately predict the service life of the blast rollers.

ObjectiveIn recent years, there has been a steady increase in the occurrence and fatality rates of shrimp diseases, causing substantial impacts in shrimp aquaculture. These diseases are marked by their swift onset, high infectivity, complex control requirements, and elevated mortality rates. With the continuous growth of shrimp factory farming, traditional manual detection approaches are no longer able to keep pace with the current requirements. Hence, there is an urgent necessity for an automated solution to identify shrimp diseases. The main goal of this research is to create a cost-effective inspection method using computer vision that achieves a harmonious balance between cost efficiency and detection accuracy. The improved YOLOv8 (You Only Look Once) network and multiple features were employed to detect shrimp diseases.MethodsTo address the issue of surface foam interference, the improved YOLOv8 network was applied to detect and extract surface shrimps as the primary focus of the image. This target detection approach accurately recognizes objects of interest in the image, determining their category and location, with extraction results surpassing those of threshold segmentation. Taking into account the cost limitations of platform computing power in practical production settings, the network was optimized by reducing parameters and computations, thereby improving detection speed and deployment efficiency. Additionally, the Farnberck optical flow method and gray level co-occurrence matrix (GLCM) were employed to capture the movement and image texture features of shrimp video clips. A dataset was created using these extracted multiple feature parameters, and a Support Vector Machine (SVM) classifier was trained to categorize the multiple feature parameters in video clips, facilitating the detection of shrimp health.Results and DiscussionsThe improved YOLOv8 in this study effectively enhanced detection accuracy without increasing the number of parameters and flops. According to the results of the ablation experiment, replacing the backbone network with FasterNet lightweight backbone network significantly reduces the number of parameters and computation, albeit at the cost of decreased accuracy. However, after integrating the efficient multi-scale attention (EMA) on the neck, the mAP0.5 increased by 0.3% compared to YOLOv8s, while mAP0.95 only decreased by 2.1%. Furthermore, the parameter count decreased by 45%, and FLOPs decreased by 42%. The improved YOLOv8 exhibits remarkable performance, ranking second only to YOLOv7 in terms of mAP0.5 and mAP0.95, with respective reductions of 0.4% and 0.6%. Additionally, it possesses a significantly reduced parameter count and FLOPS compared to YOLOv7, matching those of YOLOv5. Despite the YOLOv7-Tiny and YOLOv8-VanillaNet models boasting lower parameters and Flops, their accuracy lags behind that of the improved YOLOv8. The mAP0.5 and mAP0.95 of YOLOv7-Tiny and YOLOv8-VanillaNet are 22.4%, 36.2%, 2.3%, and 4.7% lower than that of the improved YOLOv8, respectively. Using a support vector machine (SVM) trained on a comprehensive dataset incorporating multiple feature, the classifier achieved an impressive accuracy rate of 97.625%. The 150 normal fragments and the 150 diseased fragments were randomly selected as test samples. The classifier exhibited a detection accuracy of 89% on this dataset of the 300 samples. This result indicates that the combination of features extracted using the Farnberck optical flow method and GLCM can effectively capture the distinguishing dynamics of movement speed and direction between infected and healthy shrimp. In this research, the majority of errors stem from the incorrect recognition of diseased segments as normal segments, accounting for 88.2% of the total error. These errors can be categorized into three main types: 1) The first type occurs when floating foam obstructs the water surface, resulting in a small number of shrimp being extracted from the image. 2) The second type is attributed to changes in water movement. In this study, nanotubes were used for oxygenation, leading to the generation of sprays on the water surface, which affected the movement of shrimp. 3) The third type of error is linked to video quality. When the video's pixel count is low, the difference in optical flow between diseased shrimp and normal shrimp becomes relatively small. Therefore, it is advisable to adjust the collection area based on the actual production environment and enhance video quality.ConclusionsThe multiple features introduced in this study effectively capture the movement of shrimp, and can be employed for disease detection. The improved YOLOv8 is particularly well-suited for platforms with limited computational resources and is feasible for deployment in actual production settings. However, the experiment was conducted in a factory farming environment, limiting the applicability of the method to other farming environments. Overall, this method only requires consumer-grade cameras as image acquisition equipment and has lower requirements on the detection platform, and can provide a theoretical basis and methodological support for the future application of aquatic disease detection methods.

Copyright of Journal of Chongqing University of Technology (Natural Science) is the property of Chongqing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Due to the limited availability of natural resources, it is essential that agricultural productivity keep pace with population growth. Despite unfavorable weather circumstances, this project's major objective is to boost production. As a consequence of technological advancements in agriculture, precision farming as a way for enhancing crop yields is gaining appeal and becoming more prevalent. When it comes to predicting future data, machine learning employs a number of methods, including the creation of models and the acquisition of prediction rules based on past data. In this manuscript, we examine various techniques to machine learning, as well as an automated agricultural yield projection model based on selecting the most relevant features. For the purpose of selecting features, the Grey Level Co-occurrence Matrix method is utilised. For classification, we make use of the AdaBoost Decision Tree, Artificial Neural Network (ANN), and K-Nearest Neighbour (KNN) algorithms. The data set that was used in this study is simply a compilation of information about a variety of topics, including yield, pesticide use, rainfall, and average temperature. This data collection consists of 33 characteristics or qualities in total. The crops soya beans, maze, potato, rice, paddy, wheat, and sorghum are included in this data collection. This data collection was made possible through the collaboration of the Food and Agriculture Organisation (FAO) and the World Data Bank, both of which make their data available to the public. The AdaBoost decision tree has achieved the highest level of accuracy possible when used to anticipate agricultural yield. Both the accuracy rate and the recall rate are quite high at 99 percent. Article highlights: This article presents boosting and optimum feature selection based efficient machine learning technique for accurate prediction of crop yield Proposed methodology uses Gray Level Co-occurrence Matrix technique is used to select features. The AdaBoost Decision Tree, ANN, and KNN algorithms are used for classification. The performance of AdaBoost machine learning is more accurate for crop yield prediction. Adaboost has achieved 98 percent accuracy, 99 percent Recall rate. [ABSTRACT FROM AUTHOR]

Plant disease (PD) detection is a substantial problem that needs to be tackled to develop the economy and improve agricultural production. Using conventional methods to classify plant leaf diseases consumes more time, undergoes vanishing gradients problems, overfitting issues, etc. However, automatic PD detection using deep learning (DL) has attained great significance in detecting PD during the early stages. Therefore, this paper proposes a hybrid strategy based on optimized automatic DL for plant leaf disease classification (PLDC). Initially, the proposed model performs pre-processing using image resizing and Gaussian filtering. Then, the disease infected region is then segmented using the UNet technique to acquire the relevant region and enhance disease classification accuracy. During segmentation, the weight of the UNet model has been tuned by employing the hunter-prey optimization (Hunt-PO) algorithm. Next, feature extraction is accomplished by means of a gray level co-occurrence matrix (GLCM), scale-invariant feature transform (SIFT) and a Gabor filter to extract the crucial features for classification. Further, based on the extracted features, PLDC is performed using artificial driving-EfficientNet (AD-ENet). The proposed PLDC model is implemented in the python platform through the PlantVillage dataset and assessed the performance in terms of different evaluation measures. Moreover, a proposed model's performance is compared with existing classifiers. The maximum classification accuracy obtained by the proposed PLDC model is 99.91%, superior to the existing classifiers for leaf disease classification. [ABSTRACT FROM AUTHOR]

Alzheimer's Disease (AD), the most common type of dementia, is characterized by memory issues that worsen with time. When evaluating single modalities, there are specific efficient learning techniques that produce a poor identification rate. The clinical significance of Mild Cognitive Impairment (MCI) has clinical importance, which is crucial for identifying and categorizing AD patients. This study includes the utilization of 3D T1 weighted /Magnetic Resonance Imaging (3D T1WI MRI) images where the structural changes in the brain are identified for predicting early AD. The classification accuracy for AD, MCI, and Normal Control (NC) has also increased due to cognitive evaluations when comparing three one vs one classifications, i.e. AD vs. NC, MCI vs. NC, AD vs. MCI, and multi-class classification. The proposed approach's novelty is a binary and multi-class classification of structural MRI images using a Siamese 3D Convolutional Neural Network (Siamese 3D-CNN). The Stochastic Gradient Descent (SGD) has been employed to modify the proposed network's weights, and 5-Fold Cross-Validation (CV) has been used for each binary classification task. Our study used the neuropsychological data of 417 people from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset to validate the suggested methodology. Experimental results demonstrate that accuracy, sensitivity, and specificity for three binary classifications produce the maximum percentage, in contrast to other relevant approaches found in the literature. In addition, the multi-class classification accuracy of 99.8% was attained. [ABSTRACT FROM AUTHOR]

A significant challenge for Photovoltaic (PV) power systems is the accumulation of dust on solar panels, particularly prevalent in desert areas. Dust accumulation on solar panels cause a high degradation in the output power and thus, solar panels should be monitored and cleaned continuously to keep their efficiency high. Automating the inspection of solar panels can serve as a viable alternative of human inspection due to the impact of labor expenses and human difficulties on decision-making on such an environment. In this work, we are proposing a computer vision approach that is capable of inspecting solar panels and determines its condition in terms of dust accumulation. The proposed approach aims to prove the capability of dust detection on distinct panels by means of visible light imaging and computer vision techniques. It deploys both gray level co-occurrence matrix (GLCM) textural features and local binary patterns (LBP) of solar panels' images in addition to support vector machine (SVM) to build a classification model for this purpose. The proposed approach has been tested on images of solar panels that suffer from moderate and heavy accumulation of desert sands and dusts. The experimental findings successfully illustrated the effectiveness of the proposed feature description and the overall dust detection approach of solar panels with an accuracy of 94.3%. [ABSTRACT FROM AUTHOR]

[Objective] In recent years, there has been a steady increase in the occurrence and fatality rates of shrimp diseases, causing substantial impacts in shrimp aquaculture. These diseases are marked by their swift onset, high infectivity, complex control requirements, and elevated mortality rates. With the continuous growth of shrimp factory farming, traditional manual detection approaches are no longer able to keep pace with the current requirements. Hence, there is an urgent necessity for an automated solution to identify shrimp diseases. The main goal of this research is to create a cost-effective inspection method using computer vision that achieves a harmonious balance between cost efficiency and detection accuracy. The improved YOLOv8 (You Only Look Once) network and multiple features were employed to detect shrimp diseases. [Methods] To address the issue of surface foam interference, the improved YOLOv8 network was applied to detect and extract surface shrimps as the primary focus of the image. This target detection approach accurately recognizes objects of interest in the image, determining their category and location, with extraction results surpassing those of threshold segmentation. Taking into account the cost limitations of platform computing power in practical production settings, the network was optimized by reducing parameters and computations, thereby improving detection speed and deployment efficiency. Additionally, the Farnberck optical flow method and gray level co-occurrence matrix (GLCM) were employed to capture the movement and image texture features of shrimp video clips. A dataset was created using these extracted multiple feature parameters, and a Support Vector Machine (SVM) classifier was trained to categorize the multiple feature parameters in video clips, facilitating the detection of shrimp health. [Results and Discussions] The improved YOLOv8 in this study effectively enhanced detection accuracy without increasing the number of parameters and flops. According to the results of the ablation experiment, replacing the backbone network with FasterNet lightweight backbone network significantly reduces the number of parameters and computation, albeit at the cost of decreased accuracy. However, after integrating the efficient multi-scale attention (EMA) on the neck, the mAP0.5 increased by 0.3% compared to YOLOv8s, while mAP0.95 only decreased by 2.1%. Furthermore, the parameter count decreased by 45%, and FLOPs decreased by 42%. The improved YOLOv8 exhibits remarkable performance, ranking second only to YOLOv7 in terms of mAP0.5 and mAP0.95, with respective reductions of 0.4% and 0.6%. Additionally, it possesses a significantly reduced parameter count and FLOPS compared to YOLOv7, matching those of YOLOv5. Despite the YOLOv7-Tiny and YOLOv8-VanillaNet models boasting lower parameters and Flops, their accuracy lags behind that of the improved YOLOv8. The mAP0.5 and mAP0.95 of YOLOv7-Tiny and YOLOv8-VanillaNet are 22.4%, 36.2%, 2.3%, and 4.7% lower than that of the improved YOLOv8, respectively. Using a support vector machine (SVM) trained on a comprehensive dataset incorporating multiple feature, the classifier achieved an impressive accuracy rate of 97.625%. The 150 normal fragments and the 150 diseased fragments were randomly selected as test samples. The classifier exhibited a detection accuracy of 89% on this dataset of the 300 samples. This result indicates that the combination of features extracted using the Farnberck optical flow method and GLCM can effectively capture the distinguishing dynamics of movement speed and direction between infected and healthy shrimp. In this research, the majority of errors stem from the incorrect recognition of diseased segments as normal segments, accounting for 88.2% of the total error. These errors can be categorized into three main types: 1) The first type occurs when floating foam obstructs the water surface, resulting in a small number of shrimp being extracted from the image. 2) The second type is attributed to changes in water movement. In this study, nanotubes were used for oxygenation, leading to the generation of sprays on the water surface, which affected the movement of shrimp. 3) The third type of error is linked to video quality. When the video's pixel count is low, the difference in optical flow between diseased shrimp and normal shrimp becomes relatively small. Therefore, it is advisable to adjust the collection area based on the actual production environment and enhance video quality. [Conclusions] The multiple features introduced in this study effectively capture the movement of shrimp, and can be employed for disease detection. The improved YOLOv8 is particularly well-suited for platforms with limited computational resources and is feasible for deployment in actual production settings. However, the experiment was conducted in a factory farming environment, limiting the applicability of the method to other farming environments. Overall, this method only requires consumer-grade cameras as image acquisition equipment and has lower requirements on the detection platform, and can provide a theoretical basis and methodological support for the future application of aquatic disease detection methods. [ABSTRACT FROM AUTHOR]

This study aimed to classify the maturity of strawberries and evaluate its impact on surface glossiness during storage using digital images. Five hundred images of strawberries were captured over 10 days using a digital camera. Color and texture features of the gray level co-occurrence matrix were extracted from each component of the HSV color space, which were used to classify maturity levels through linear discriminant analysis. The strawberry samples were classified with an overall accuracy of 82% in five classes. Furthermore, as an index of surface glossiness to reflect visual freshness, changes in the luminance distribution on the strawberry surface were evaluated. The later the harvest, the quicker the deterioration in the glossiness of strawberries. Strawberries that are ripe enough for harvesting but still relatively immature can be stored for a longer period than fully ripe strawberries while retaining better surface glossiness. [ABSTRACT FROM AUTHOR]

In view of the low detection rate of steel surface defect classification, a surface defect classification method based on texture multi-scale feature fusion is adopted. Gabor filter and gray level co-occurrence matrix are used to establish multi-scale feature vectors of texture images. At the same time, convolution operation is used to extract features of texture images, and hybrid dilated convolution module is introduced to increase receptive field. The two feature vectors are fused to obtain the enhanced fused texture feature vector. The fused features are input into Long Short-Term Memory(LSTM) network in sequence to build a classification model, and the classification results are evaluated using the confusion matrix. The results show that the classification accuracy of this method on the NEU(Northeastern University) data set is 97.5%. The LSTM network, BP(Back Propagation) neural network, SVM(Support Vector Machine), KNN(K-Nearest Neighbor), CART(Classification And Regression Tree) and other classification methods are set up for comparative experiments. The results show that LSTM classification method performs best in multi-scale, and F1 index is the highest. Ablation experiments are conducted with BP network, LSTM network, SVM, KNN, CART, CNN, AlexNet and other methods to verify the universality of this method. This method fully exploits the multi-scale feature information of the texture image, and has important significance for the research of the classification method of steel surface defects. [ABSTRACT FROM AUTHOR]

Copyright of Food & Machinery is the property of Food & Machinery Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Skin Cancer is the most common form of the disease and is responsible for millions of deaths each year. Most of the relevant studies concentrate on algorithms that are based on machine learning, and few on deep learning as well. However, due to the several challenges in dermoscopic image acquisition, these algorithms are unable to deliver the highest possible level of accuracy and specificity. Therefore, this article implements skin cancer detection and classification (SCDC) system using multilevel feature extraction (MFE)-based artificial intelligence (AI) with unsupervised learning (USL), here after denoted as MFEUsLNet. Initially, the given skin images are preprocessed using bilateral filter, which removes the noise artifacts from the source images. Then, a well-known USL approach named K-means clustering (KMC) is used for segmentation of skin lesion, which can detect the affected skin lesion quite efficiently. Then, gray level co-occurrence matrix (GLCM), and redundant discrete wavelet transform (RDWT) are used for low level, texture and colour feature extraction. Finally, recurrent neural network (RNN) classifier is used to train with these multi-level features and classify the multiple types of skin cancer. The simulations proven that the proposed MFEUsLNet model is outperformed state-of-the-art SCDC approaches in terms of medical statistical quality metrics such as classification accuracy, specificity, precision, recall, F1-score, and sensitivity for ISIC-2020 dataset.

Non-Destructive Testing (NDT) is important to detect sub-surface defects in the weldments to ensure the quality of weld joints. The weld radiographs are digitized using a high-resolution digital camera. Data augmentation techniques are applied to expand the radiographic image dataset. Multi-class defect classification is done using the Gray-level co-occurrence matrix as a feature extractor and these features are given as input to various classifiers for classifying slag inclusion, incomplete penetration, and acceptable weld bead classes. The proposed methodology achieved the highest accuracies of 84%,83%,80%,70%, and 64% respectively for GLCM plus Random Forest, GLCM plus XGBoost, GLCM plus lightGBM, GLCM plus KNN, and GLCM plus SVM. The technology of applying ML techniques on radiographic images in detection of defects in welding as well as other manufacturing processes can be a sustainable practice.

In most data dimensionality reduction tasks, feature selection (FS) is an essential preprocessing stage. The slime mould algorithm (SMA) is a novel metaheuristic optimization algorithm. This paper proposes a feature selection method for banknote dirtiness recognition based on mathematical functions driven slime mould algorithm. This method is applied to the feature selection of banknote dirtiness recognition, and the Lévy flight operator is used to replace the control parameter vc in SMA, and a series of mathematical functions are added to replace the control parameter vb in the slime mould algorithm to enhance the global search capability. Test results show that this method achieves the highest accuracy of 76.85 % in recognizing the banknote dirtiness. Additionally, by applying the same method to feature selection and recognition of dirtiness for the left middle white area images of banknotes, the highest accuracy reaches 74.99 %. This suggests that the left middle white area can essentially replace the whole banknote. To verify the performance of the proposed feature selection method, comparative tests with four other optimization algorithms are carried out, and the results demonstrate that three improvement strategies can effectively improve the performance of SMA while maintaining a balance between exploration and exploitation. Tests have revealed that these three algorithms have significant advantages over the past SMA in terms of recognition accuracy and convergence speed when it comes to identifying the dirtiness of banknotes. Therefore, this method is considered to be one of the most effective methods to solve the feature selection problem of banknote dirtiness. [ABSTRACT FROM AUTHOR]

This work aimed to compare the discrimination ability and detection of limit on the different early implicit bruised pears based on visible and short-wave near-infrared (400–1000 nm) hyperspectral imaging (Vis-SWNIR-HSI) features and spectral information. The changes of physical characteristics were much more sensitive than chemical parameters for bruised pears while the postharvest storage. The pear bruising changes can be addressed to the spectral variations at 416–486 nm, 516–646 nm, 939–954 nm, and 973 nm. The partial least squares discrimination models (PLS-DA) based on selected HSI spectral variables can satisfactorily identify the bruised pear area about 63.32 ± 9.02 mm2, with the discrimination accuracy of 90%, but not available of using imaging features. Further, PLS-DA models based on the fusion of selected HSI spectral variables and imaging features reached the discrimination rate of 98.8% between bruised and non-bruised pears, and provided the discrimination rate of 95% to successfully identify the early bruised pears with bruise area of 72.18 ± 7.32 mm2 after 1 day of storage. Consequently, the combination of hyperspectral imaging features and spectral variables had the best ability to detect early implicit bruised 'Korla' fragrant pears during postharvest storage. • Early bruised 'Korla' pears showed significant changes in firmness and flesh color. • Bruised pears can be detected by HSI spectral information with accuracy of 96.7%. • Lower detection accuracy of bruised pears used HSI imaging features than spectrum. • The HSI detection limit of pear bruised area can reach about 63.32 ± 9.02 mm2. [ABSTRACT FROM AUTHOR]