Search

Video-based Point Cloud Compression (V-PCC) is a dynamic point cloud coding standard that compresses 3D point clouds by projecting them into 2D frames, involving computationally intensive steps that can hinder real-time applications in augmented and virtual reality. Several studies have endeavoured to mitigate this challenge by employing voxel classification and adaptive selection of voxels, yet the computationally intensive patch generation and refinement procedures persist. This paper presents an innovative approach that utilizes hierarchical hexahedron partitioning, adapting to varying point cloud densities and non-planar surfaces based on transmission requirements. The aim is to enhance projection by directly converting 3D data into 2D frames, bypassing segmentation and refinement. Crucially, the proposed method improves data capture by fine-tuning the projection layer of 2D frames, surpassing the capabilities of V-PCC’s projection layer, and effectively addressing potential patch-boundary discontinuities. Furthermore, this approach enhances the efficiency of 2D frames by producing smaller maps than V-PCC. Experimental results demonstrate the efficacy of the proposed method by significantly reducing patch generation time in V-PCC while preserving quality and achieving significant bitrate savings. By employing the proposed method, it is achievable to attain approximately a 7.32% improvement in geometry BD-Rate (D1) and 6.61% (D2), along with a notable reduction in time complexity ranging from 33% to 59% based on high/low bitrate scenarios compared to the V-PCC anchor. This research thus tackles the pressing challenges of V-PCC by proposing a promising solution for the real-time application of dynamic point cloud compression.

Recent years have encountered a noticeable expansion of point cloud-based 3D applications that compels the necessity of high-efficiency point cloud compression. Preserving the local density of point cloud is crucial for compression, however, it’s challenging due to the unordered nature of points in 3D space and has been overlooked by the majority of the existing compression methods. Recently, the farthest point method has demonstrated effectiveness in sampling point clouds across various state-of-the-art methods, its limitation lies in being non-density aware, yielding reconstructions that fall short of the desired quality. Intending to achieve density-aware compression for improved reconstruction, this paper proposes an end-to-end learnable cluster-based compression method for efficient lossy point cloud geometry compression. Our method utilizes an autoencoder architecture that performs point-wise operation for compression and reconstruction. To extract effective latent features, the encoder segments the point clouds into clusters using a different approach than farthest point sampling that is capable of well capturing uneven point densities and employs cluster-wise compression independently. To enhance density retention further, we leveraged the capabilities of an attention mechanism, allowing it to learn complex point-wise dependencies within clusters and effectively capture local density information. At the decoder, decompressed clusters are accumulated to reconstruct point clouds completely. In terms of rate-distortion trade-off, the experimental analysis reveals the superiority of the proposed method over prior arts. Furthermore, our approach adapts well to different datasets, for example, the model learned from the ModelNet40 dataset works well and achieves state-of-the-art performances on ShapeNet datasets. Finally, the qualitative comparison demonstrates that the proposed framework can preserve satisfactory local geometry details of point clouds compared to existing Representative methods.

A point cloud is a representation of objects or scenes utilising unordered points comprising 3D positions and attributes. The ability of point clouds to mimic natural forms has gained significant attention from diverse applied fields, such as virtual reality and augmented reality. However, the point cloud, especially those representing dynamic scenes or objects in motion, must be compressed efficiently due to its huge data volume. The latest video-based point cloud compression (V-PCC) standard for dynamic point clouds divides the 3D point cloud into many patches using computationally expensive normal estimation, segmentation, and refinement. The patches are projected onto a 2D plane to apply existing video coding techniques. This process often results in losing proximity information and some original points. This loss induces artefacts that adversely affect user perception. The proposed method segments dynamic point clouds based on shape similarity and occlusion before patch generation. This segmentation strategy helps maintain the points’ proximity and retain more original points by exploiting the density and occlusion of the points. The experimental results establish that the proposed method significantly outperforms the V-PCC standard and other relevant methods regarding rate–distortion performance and subjective quality testing for both geometric and texture data of several benchmark video sequences.

Abstract In recent years, quantum image computing draws a lot of attention due to storing and processing image data faster compared to classical computers. A number of approaches have been proposed to represent the quantum image inside a quantum computer. Representing and compressing medium and big-size images inside the quantum computer is still challenging. To address this issue, we have proposed a block-wise DCT-EFRQI (Direct Cosine Transform Efficient Flexible Representation of Quantum Image) approach to represent and compress the gray-scale image efficiently to save computational time and reduce the quantum bits (qubits) for the state preparation. In this work, we have demonstrated the capability of block-wise DCT and DWT transformation inside the quantum domain to investigate their relative performances. The Quirk simulation tool is used to design the corresponding quantum image circuit. In the proposed DCT-EFRQI approach, a total of 17 qubits are used to represent the coefficients, the connection between coefficients and state (i.e., auxiliary), and their position for representing and compressing grayscale images inside a quantum computer. Among those, 8 qubits are used to map the coefficient values and the rest are used to generate the corresponding coefficient XY-coordinate position including one auxiliary qubit. Theoretical analysis and experimental results show that the proposed DCT-EFRQI scheme provides better representation and compression compared to DCT-GQIR, DWT-GQIR, and DWT-EFRQI in terms of rate-distortion performance.

Lung cancer is the leading cause of cancer death, and early diagnosis is associated with a positive prognosis. Chest X-ray (CXR) provides an inexpensive imaging mode for lung cancer diagnosis. Computer vision algorithms have previously been proposed to assist human radiologists in this task; however, leading studies use down-sampled images and computationally expensive methods with unproven generalization. In contrast, this study localizes lung nodules from CXR images using efficient encoder-decoder neural networks that have been crafted to process full resolution input images, thereby avoiding signal loss resulting from down-sampling. Encoder-decoder networks are trained and tested using the Japanese Society of Radiological Technology dataset. The networks are used to localize lung nodules from an independent CXR dataset. These experiments allow for the determination of the optimal network depth, image resolution, and pre-processing pipeline for generalized lung nodule localization. We find that more subtle nodules are detected in earlier training epochs. Therefore, we propose a novel self-ensemble model from three consecutive epochs centered on the validation optimum. This ensemble achieved a sensitivity of 85% in 10-fold internal testing with false positives of 8 per image. A sensitivity of 81% is achieved at a false positive rate of 6 following morphological false positive reduction. This result is comparable to more computationally complex systems, but with a sub-second inference time that is faster than other methods presented in the literature. The proposed algorithm achieved excellent generalization results against a challenging external dataset with a sensitivity of 77% at a false positive rate of 7.6.

Lane marking detection is fundamental for both advanced driving assistance systems and traffic surveillance systems. However, detecting lane is highly challenging when the visibility of a road lane marking is low, obscured or often invisible due to real-life challenging environment and adverse weather. Most of the lane detection methods suffer from four types of challenges: (i) light effects i.e. shadow, glare of light, reflection etc. created by different light sources like streetlamp, tunnel-light, sun, wet road etc.; (ii) Obscured visibility of eroded, blurred, dashed, colored and cracked lane caused by natural disasters and adverse weather (rain, snow etc.); (iii) lane marking occlusion by different objects from surroundings (wiper, vehicles etc.); and (iv) presence of confusing lane like lines inside the lane view e.g., guardrails, pavement marking, road divider etc. In this paper, we proposed a simple, real-time, and robust lane detection and tracking method to detect lane marking considering the abovementioned challenging conditions. In this method, we introduced three key technologies. First, we introduce a comprehensive intensity threshold range (CITR) to improve the performance of the canny operator in detecting different types of lane edges e.g., clear, low intensity, cracked, colored, eroded, or blurred lane edges. Second, we propose a two-step lane verification technique, the angle-based geometric constraint (AGC) and length-based geometric constraint (LGC) followed by Hough Transform, to verify the characteristics of lane marking and to prevent incorrect lane detection. Finally, we propose a novel lane tracking technique, to predict the lane position of the next frame by defining a range of horizontal lane position (RHLP) along the x axis which will be updated with respect to the lane position of previous frame. It can keep track of the lane position when either left or right or both lane markings are partially and fully invisible. To evaluate the performance of the proposed method we used the DSDLDE (Lee and Moon, 2018) and SLD (Borkar et al., 2009) dataset with $1080\times 1920$ and $480\times 720$ resolutions at 24 and 25 frames/sec respectively where the video frames containing different challenging scenarios. Experimental results show that the average detection rate is 97.55%, and the average processing time is 22.33 msec/frame, which outperforms the state-of-the-art method.

Abstract In recent years, the nuclear power plant has received huge attention as it generates vast amounts of power at a lower cost. However, its creation of radioactive wastes is a major environmental concern. Therefore, the nuclear power plant requires a reliable and uninterrupted monitoring system as an essential part of it. Monitoring a nuclear power plant using wireless sensor networks is a convenient and popular practice now. This paper proposes a hybrid approach for monitoring wireless sensor networks in the context of a nuclear power plant in Bangladesh. Our hybrid approach enhances the lifespan of wireless sensor networks reducing power consumption and offering better connectivity of sensors. To do so, it uses both the topology maintenance and topology construction algorithms. We found that the HGETRecRot topology maintenance algorithm enhances the network lifetime compared to other algorithms. This algorithm increases the communication and sensing coverage area but decreases the network performance. We also propose a prediction model, based on linear regression algorithm, that predicts the best combination of topology maintenance and topology construction algorithms.

Equipment for the correct identification of living objects entrapped under heavy debris is generally purpose-built, is costly, must be operated by highly trained professionals and is not readily available in a catastrophic event. A more readily available solution for improving the time-to-rescue ratio and logistics issues can be provided with smartphones which, equipped with software to find signs of life, are readily available at any disaster scene. This paper examines whether cardiac and pulmonary-related activities of living objects can provide acceptably accurate readings from a non-contact detection method. Laboratory experiments were conducted with Doppler radar at a 2.4 GHz frequency spectrum similar to smartphone-like devices, with empirical results demonstrating that human vital signs can be clearly identified when using smartphones for non-contact detection of living objects entrapped under debris. Experiments also simulated the psychogenic tremors likely to be experienced by individuals while operating the sensor-equipped devices under crisis conditions. The results show a clear relationship between the wavelength of pulmonary and blood vessel activities and the distance between the trapped human and the sensor in various conditions. The article also reports the design of a pseudo learning algorithm for model-based anomaly detection in time series to detect vital signs during normal and abnormal ventilation based on cardiopulmonary clinical records and datasets. This work significantly contributes to the existing body of research on timely rescue during disaster events.

The Region of interest (ROI) analysis is widely used in image analytics, video coding, computer graphics, computer vision, medical imaging, nuclear medicine, computer tomography and many other areas in medical applications. This ROI determination process using subjective method (e.g. using human vision) often differ from the objective ones (e.g. using mathematical modelling). However, there is no existing method in the literature that could provide a single decision when both methods’ ROI data is available. To address this limitation, a robust algorithm is developed by combining the human eye tracking (subjective) and the graph-based visual saliency modelling (objective) information to determine a more realistic ROI for a scene. To carry out this process, in one hand, several different independent human visual saliency factors such as pupil size, pupil dilation, central tendency, fixation pattern, and gaze plot for a group of twenty-two participants are collected by applying on a set of publicly available eighteen video sequences. On the other hand, the features of Graph based visual saliency (GBVS) highlights conspicuity in the scene. Gleaned from these two pieces of information, the proposed algorithm determines the final ROI based on some heuristics. Experimental results show that for a wide range of video sequences and compared to the existing deep learning based (MxSalNet) and depth pixel (DP) based ROI, the proposed ROI is more consistent to the benchmark ROI, which was previously decided by a group of video coding experts. As the subjective and objective options frequently create an ambiguity to reach a single decision on ROI, the proposed algorithm could determine an ultimate decision, which is eventually validated by experts’ opinion.

In computer vision applications, the visibility of the video content is crucial to perform analysis for better accuracy. The visibility can be affected by several atmospheric interferences in challenging weather – one such interference is the appearance of rain streaks. Recently, rain streak removal has achieved plenty of interest among researchers, as it has some exciting applications such as autonomous cars, intelligent traffic monitoring systems, multimedia, etc. In this paper, we propose a novel and simple method of rain streak removal by combining three novel extracted visual features focusing on the temporal appearance, wide shape and relative location of the rain streak. We called it the TAWL (Temporal Appearance, Width, and Location) method. The proposed TAWL method adaptively uses features from different resolutions and frame rates. Moreover, it progressively processes features from the upcoming frames so that it can remove rain in real-time. Experiments have been conducted using video sequences with both real rain and synthetic rain to compare the performance of the proposed method against the relevant state-of-the-art methods. The experimental results demonstrate that the proposed method outperforms the state-of-the-art methods by removing more rain streaks while keeping other moving regions.

The COVID-19 pandemic has inspired unprecedented data collection and computer vision modelling efforts worldwide, focused on the diagnosis of COVID-19 from medical images. However, these models have found limited, if any, clinical application due in part to unproven generalization to data sets beyond their source training corpus. This study investigates the generalizability of deep learning models using publicly available COVID-19 Computed Tomography data through cross dataset validation. The predictive ability of these models for COVID-19 severity is assessed using an independent dataset that is stratified for COVID-19 lung involvement. Each inter-dataset study is performed using histogram equalization, and contrast limited adaptive histogram equalization with and without a learning Gabor filter. We show that under certain conditions, deep learning models can generalize well to an external dataset with F1 scores up to 86%. The best performing model shows predictive accuracy of between 75% and 96% for lung involvement scoring against an external expertly stratified dataset. From these results we identify key factors promoting deep learning generalization, being primarily the uniform acquisition of training images, and secondly diversity in CT slice position.

Screening programs for early lung cancer diagnosis are uncommon, primarily due to the challenge of reaching at-risk patients located in rural areas far from medical facilities. To overcome this obstacle, a comprehensive approach is needed that combines mobility, low cost, speed, accuracy, and privacy. One potential solution lies in combining the chest X-ray imaging mode with federated deep learning, ensuring that no single data source can bias the model adversely. This study presents a pre-processing pipeline designed to debias chest X-ray images, thereby enhancing internal classification and external generalization. The pipeline employs a pruning mechanism to train a deep learning model for nodule detection, utilizing the most informative images from a publicly available lung nodule X-ray dataset. Histogram equalization is used to remove systematic differences in image brightness and contrast. Model training is then performed using combinations of lung field segmentation, close cropping, and rib/bone suppression. The resulting deep learning models, generated through this pre-processing pipeline, demonstrate successful generalization on an independent lung nodule dataset. By eliminating confounding variables in chest X-ray images and suppressing signal noise from the bone structures, the proposed deep learning lung nodule detection algorithm achieves an external generalization accuracy of 89%. This approach paves the way for the development of a low-cost and accessible deep learning-based clinical system for lung cancer screening.

Video coding using dynamic background frame achieves better compression compared to the traditional techniques by encoding background and foreground separately. This process reduces coding bits for the overall frame significantly; however, encoding background still requires many bits that can be compressed further for achieving better coding efficiency. The cuboid coding framework has been proven to be one of the most effective methods of image compression which exploits homogeneous pixel correlation within a frame and has better alignment with object boundary compared to traditional block-based coding. In a video sequence, the cuboid-based frame partitioning varies with the changes of the foreground. However, since the background remains static for a group of pictures, the cuboid coding exploits better spatial pixel homogeneity. In this work, the impact of cuboid coding on the background frame for high-resolution videos (Ultra-High-Definition (UHD) and 360-degree videos) is investigated using the multilayer framework of SHVC. After the cuboid partitioning, the method of coarse frame generation has been improved with a novel idea by keeping human-visual sensitive information. Unlike the traditional SHVC scheme, in the proposed method, cuboid coded background and the foreground are encoded in separate layers in an implicit manner. Simulation results show that the proposed video coding method achieves an average BD-Rate reduction of 26.69% and BD-PSNR gain of 1.51 dB against SHVC with significant encoding time reduction for both UHD and 360 videos. It also achieves an average of 13.88% BD-Rate reduction and 0.78 dB BD-PSNR gain compared to the existing relevant method proposed by X. Hoang Van.

Due to the availability of powerful image-editing software and the growing amount of multimedia data that is transmitted via the Internet, integrity verifications and confidentiality of the data are becoming critical issues. However, currently, the accuracy of detecting and the recovery capability of the tampered images by the existing methods through watermarking strategy is still not at the required level, especially at a higher tampered rate. This paper proposes a new blind and fragile watermarking method to detect tampering and better recovery of tampered images. To improve the quality of both the watermarked and the recovered images, a new feature extraction scheme is introduced which will produce a short but comprehensive recovery code using a new compression strategy. If a block in the image tampers, the proposed embedded feature allows the original data to be extracted for recovery. To overcome tamper coincidence, every block’s watermarked data contains not only the recovery code belonging to the block itself but also its neighbor’s data as a second layer of recovery. Various size blocks were investigated to see the performance and compare their efficiency for recovering an image after different tampering rates. The test showed the smaller block sizes may be more suitable for locating tampering, where the bigger ones are more suitable when the tampering rate is higher. The bigger block sizes in the proposed method can recover an image even after a 60% tampering rate with high quality (more than 31 dB). The experimental results prove that the proposed method can have better efficiency for detecting tampering, and recovery of the original image, compared to the relevant existing methods.

Detecting COVID-19 early may help in devising an appropriate treatment plan and disease containment decisions. In this study, we demonstrate how transfer learning from deep learning models can be used to perform COVID-19 detection using images from three most commonly used medical imaging modes X-Ray, Ultrasound, and CT scan. The aim is to provide over-stressed medical professionals a second pair of eyes through intelligent deep learning image classification models. We identify a suitable Convolutional Neural Network (CNN) model through initial comparative study of several popular CNN models. We then optimize the selected VGG19 model for the image modalities to show how the models can be used for the highly scarce and challenging COVID-19 datasets. We highlight the challenges (including dataset size and quality) in utilizing current publicly available COVID-19 datasets for developing useful deep learning models and how it adversely impacts the trainability of complex models. We also propose an image pre-processing stage to create a trustworthy image dataset for developing and testing the deep learning models. The new approach is aimed to reduce unwanted noise from the images so that deep learning models can focus on detecting diseases with specific features from them. Our results indicate that Ultrasound images provide superior detection accuracy compared to X-Ray and CT scans. The experimental results highlight that with limited data, most of the deeper networks struggle to train well and provides less consistency over the three imaging modes we are using. The selected VGG19 model, which is then extensively tuned with appropriate parameters, performs in considerable levels of COVID-19 detection against pneumonia or normal for all three lung image modes with the precision of up to 86% for X-Ray, 100% for Ultrasound and 84% for CT scans.

The COVID-19 pandemic has triggered an urgent call to contribute to the fight against an immense threat to the human population. Computer Vision, as a subfield of artificial intelligence, has enjoyed recent success in solving various complex problems in health care and has the potential to contribute to the fight of controlling COVID-19. In response to this call, computer vision researchers are putting their knowledge base at test to devise effective ways to counter COVID-19 challenge and serve the global community. New contributions are being shared with every passing day. It motivated us to review the recent work, collect information about available research resources, and an indication of future research directions. We want to make it possible for computer vision researchers to find existing and future research directions. This survey article presents a preliminary review of the literature on research community efforts against COVID-19 pandemic.

Soil colour is one of the most important factors in agriculture for monitoring soil health and determining its properties. For this purpose, Munsell soil colour charts are widely used by archaeologists, scientists, and farmers. The process of determining soil colour from the chart is subjective and error-prone. In this study, we used popular smartphones to capture soil colours from images in the Munsell Soil Colour Book (MSCB) to determine the colour digitally. These captured soil colours are then compared with the true colour determined using a commonly used sensor (Nix Pro-2). We have observed that there are colour reading discrepancies between smartphone and Nix Pro-provided readings. To address this issue, we investigated different colour models and finally introduced a colour-intensity relationship between the images captured by Nix Pro and smartphones by exploring different distance functions. Thus, the aim of this study is to determine the Munsell soil colour accurately from the MSCB by adjusting the pixel intensity of the smartphone-captured images. Without any adjustment when the accuracy of individual Munsell soil colour determination is only 9% for the top 5 predictions, the accuracy of the proposed method is 74%, which is significant.

Light detection and ranging (LiDAR) sensors have accrued an ever-increasing presence in the agricultural sector due to their non-destructive mode of capturing data. LiDAR sensors emit pulsed light waves that return to the sensor upon bouncing off surrounding objects. The distances that the pulses travel are calculated by measuring the time for all pulses to return to the source. There are many reported applications of the data obtained from LiDAR in agricultural sectors. LiDAR sensors are widely used to measure agricultural landscaping and topography and the structural characteristics of trees such as leaf area index and canopy volume; they are also used for crop biomass estimation, phenotype characterisation, crop growth, etc. A LiDAR-based system and LiDAR data can also be used to measure spray drift and detect soil properties. It has also been proposed in the literature that crop damage detection and yield prediction can also be obtained with LiDAR data. This review focuses on different LiDAR-based system applications and data obtained from LiDAR in agricultural sectors. Comparisons of aspects of LiDAR data in different agricultural applications are also provided. Furthermore, future research directions based on this emerging technology are also presented in this review.

Soil moisture, soil organic carbon, and nitrogen content prediction are considered significant fields of study as they are directly related to plant health and food production. Direct estimation of these soil properties with traditional methods, for example, the oven-drying technique and chemical analysis, is a time and resource-consuming approach and can predict only smaller areas. With the significant development of remote sensing and hyperspectral (HS) imaging technologies, soil moisture, carbon, and nitrogen can be estimated over vast areas. This paper presents a generalized approach to predicting three different essential soil contents using a comprehensive study of various machine learning (ML) models by considering the dimensional reduction in feature spaces. In this study, we have used three popular benchmark HS datasets captured in Germany and Sweden. The efficacy of different ML algorithms is evaluated to predict soil content, and significant improvement is obtained when a specific range of bands is selected. The performance of ML models is further improved by applying principal component analysis (PCA), a dimensional reduction method that works with an unsupervised learning method. The effect of soil temperature on soil moisture prediction is evaluated in this study, and the results show that when the soil temperature is considered with the HS band, the soil moisture prediction accuracy does not improve. However, the combined effect of band selection and feature transformation using PCA significantly enhances the prediction accuracy for soil moisture, carbon, and nitrogen content. This study represents a comprehensive analysis of a wide range of established ML regression models using data preprocessing, effective band selection, and data dimension reduction and attempt to understand which feature combinations provide the best accuracy. The outcomes of several ML models are verified with validation techniques and the best- and worst-case scenarios in terms of soil content are noted. The proposed approach outperforms existing estimation techniques.

As the human vision is a definitive assessor of video quality, the expanded interest for no-reference subjective quality assessment (SQA) is focusing on a definitive goal of coordinating with human observation. However, the widely used subjective estimator-mean opinion score (MOS) is often biased by the testing environment, viewers mode, expertise, domain knowledge, and other factors which may influence on actual assessment. In this paper, a no-reference SQA metric is devised by simply exploiting the nature of human eye browsing on videos and analyzing the associated quality correlation features. The high efficiency video coding (HEVC) reference test model is first employed to produce different forms of coded video quality which then displayed to a number of partakers. Their eye-tracker recorded spatiotemporal gaze-data indicate more concentrated eye-traversing approach for relatively better quality. Thus, we calculate the quality assessment related to assorted features such as length pursuit, angle deflection, pupil deviation, and gaze interlude from recorded gaze trajectory. The content and resolution invariant operations are carried out prior to synthesizing them using an adaptive weighted function to develop a new quality metric-eye maneuver (EMAN). Tested results reveal that the quality evaluation carried out by the EMAN is comparatively better than MOS and structural similarity (SSIM) in terms of assessing different aspects of coded video quality for a wide range of single view video contents. For the free viewpoint video (FVV), where the reference frame is not available, the EMAN could also better distinguish different qualities compared to the MOS and SSIM.

This paper presents a comprehensive review of the principle and application of deep learning in retinal image analysis. Many eye diseases often lead to blindness in the absence of proper clinical diagnosis and medical treatment. For example, diabetic retinopathy (DR) is one such disease in which the retinal blood vessels of human eyes are damaged. The ophthalmologists diagnose DR based on their professional knowledge, that is labor intensive. With the advances in image processing and artificial intelligence, computer vision-based techniques have been applied rapidly and widely in the field of medical images analysis and are becoming a better way to advance ophthalmology in practice. Such approaches utilize accurate visual analysis to identify the abnormality of blood vessels with improved performance over manual procedures. More recently, machine learning, in particular, deep learning, has been successfully implemented in this area. In this paper, we focus on recent advances in deep learning methods for retinal image analysis. We review the related publications since 1982, which include more than 80 papers for retinal vessels detections in the research scope spanning from segmentation to classification. Although deep learning has been successfully implemented in other areas, we found only 17 papers so far focus on retinal blood vessel segmentation. This paper characterizes each deep learning based segmentation method as described in the literature. Analyzing along with the limitations and advantages of each method. In the end, we offer some recommendations for future improvement for retinal image analysis.

An automatic retinal vessel segmentation method is proposed for quick and accurate segmentation of eye vessels. Such a method would play a vital role in analyzing many eye diseases. A retinal fundus image contains varying low contrasts, which undermine the performance of the segmentation process. Independent component analysis (ICA) is largely used for noise removal and consists of two architectures, designated as ICA1 and ICA2. We have validated both ICA architectures on retinal color fundus images and selected the one that provides improved contrast values. For retinal fundus, ICA2 architecture performed better than ICA1 by virtue of being more effective in compensating the low contrast values. Experiments conducted here validated the improvements over previously reported state-of-the-art methods. The impact of proposed segmentation model was assessed on publicly available databases like DRIVE and STARE. In case of the DRIVE database, the sensitivity increased 3% by (from 72% to 75%) while maintaining a segmentation accuracy of around 96%.

The latest high efficiency video coding (HEVC) standard introduces a large number of inter-mode block partitioning modes. The HEVC reference test model (HM) uses partially exhaustive tree-structured mode selection, which still explores a large number of prediction unit (PU) modes for a coding unit (CU). This impacts on encoding time rise which deprives a number of electronic devices having limited processing resources to use various features of HEVC. By analyzing the homogeneity, residual, and different statistical correlation among modes, many researchers speed-up the encoding process through the number of PU mode reduction. However, these approaches could not demonstrate the similar rate-distortion (RD) performance with the HM due to their dependency on existing Lagrangian cost function (LCF) within the HEVC framework. In this paper, to avoid the complete dependency on LCF in the initial phase, we exploit visual sensitive foreground motion and spatial salient metric (FMSSM) in a block. To capture its motion and saliency features, we use the dynamic background and visual saliency modeling, respectively. According to the FMSSM values, a subset of PU modes is then explored for encoding the CU. This preprocessing phase is independent from the existing LCF. As the proposed coding technique further reduces the number of PU modes using two simple criteria (i.e., motion and saliency), it outperforms the HM in terms of encoding time reduction. As it also encodes the uncovered and static background areas using the dynamic background frame as a substituted reference frame, it does not sacrifice quality. Tested results reveal that the proposed method achieves 32% average encoding time reduction of the HM without any quality loss for a wide range of videos.

Video analytics and computer vision applications face challenges when using video sequences with low visibility. The visibility of a video sequence is degraded when the sequence is affected by atmospheric interference like rain. Many approaches have been proposed to remove rain streaks from video sequences. Some approaches are based on physical features, and some are based on data-driven (i.e., deep-learning) models. Although the physical features-based approaches have better rain interpretability, the challenges are extracting the appropriate features and fusing them for meaningful rain removal, as the rain streaks and moving objects have dynamic physical characteristics and are difficult to distinguish. Additionally, the outcome of the data-driven models mostly depends on variations relating to the training dataset. It is difficult to include datasets with all possible variations in model training. This paper addresses both issues and proposes a novel hybrid technique where we extract novel physical features and data-driven features and then combine them to create an effective rain-streak removal strategy. The performance of the proposed algorithm has been tested in comparison to several relevant and contemporary methods using benchmark datasets. The experimental result shows that the proposed method outperforms the other methods in terms of subjective, objective, and object detection comparisons for both synthetic and real rain scenarios by removing rain streaks and retaining the moving objects more effectively.

Lung cancer is the leading cause of cancer death and morbidity worldwide. Many studies have shown machine learning models to be effective in detecting lung nodules from chest X-ray images. However, these techniques have yet to be embraced by the medical community due to several practical, ethical, and regulatory constraints stemming from the “black-box” nature of deep learning models. Additionally, most lung nodules visible on chest X-rays are benign; therefore, the narrow task of computer vision-based lung nodule detection cannot be equated to automated lung cancer detection. Addressing both concerns, this study introduces a novel hybrid deep learning and decision tree-based computer vision model, which presents lung cancer malignancy predictions as interpretable decision trees. The deep learning component of this process is trained using a large publicly available dataset on pathological biomarkers associated with lung cancer. These models are then used to inference biomarker scores for chest X-ray images from two independent data sets, for which malignancy metadata is available. Next, multi-variate predictive models were mined by fitting shallow decision trees to the malignancy stratified datasets and interrogating a range of metrics to determine the best model. The best decision tree model achieved sensitivity and specificity of 86.7% and 80.0%, respectively, with a positive predictive value of 92.9%. Decision trees mined using this method may be considered as a starting point for refinement into clinically useful multi-variate lung cancer malignancy models for implementation as a workflow augmentation tool to improve the efficiency of human radiologists.

The efficiency of a vineyard management system is directly related to the effective management of nutritional disorders, which significantly downgrades vine growth, crop yield and wine quality. To detect nutritional disorders, we successfully extracted a wide range of features using hyperspectral (HS) images to identify healthy and individual nutrient deficiencies of grapevine leaves. Features such as mean reflectance, mean first derivative reflectance, variation index, mean spectral ratio, normalised difference vegetation index (NDVI) and standard deviation (SD) were employed at various stages in the ultraviolet (UV), visible (VIS) and near-infrared (N.I.R.) regions for our experiment. Leaves were examined visually in the laboratory and grouped as either healthy (i.e. control) or unhealthy. Then, the features of the leaves were extracted from these two groups. In a second experiment, features of individual nutrient-deficient leaves (e.g., N, K and Mg) were also analysed and compared with those of control leaves. Furthermore, a customised support vector machine (SVM) was used to demonstrate that these features can be utilised with a high degree of effectiveness to identify unhealthy samples and not only to distinguish from control and nutrient deficient but also to identify individual nutrient defects. Therefore, the proposed work corroborated that HS imaging has excellent potential to analyse features based on healthiness and individual nutrient deficiencies of grapevine leaves.

Detecting animals to estimate abundance can be difficult, particularly when the habitat is dense or the target animals are fossorial. The recent surge in the use of thermal imagers in ecology and their use in animal detections can increase the accuracy of population estimates and improve the subsequent implementation of management programs. However, the use of thermal imagers results in many hours of captured flight videos which require manual review for confirmation of species detection and identification. Therefore, the perceived cost and efficiency trade-off often restricts the use of these systems. Additionally, for many off-the-shelf systems, the exported imagery can be quite low resolution (

Video summarization is a process to extract objects and their activities from a video and represent them in a condensed form. Existing methods for video summarization fail to detect moving (dynamic) objects in the low color contrast area of a video frame due to the pixel intensities of objects and non-objects are almost similar. However, edges of objects are prominent in the low contrast regions. Moreover, to represent objects, geometric primitives (such as lines, arcs) are distinguishable and high level shape descriptors than edges. In this paper, a novel method is proposed for video summarization using geometric primitives such as conic parts, line segments and angles. Using these features, objects are extracted from each video frame. A cost function is applied to measure the dissimilarity of locations of geometric primitives to detect the movement of objects between consecutive frames. The total distance of object movement is calculated and each video frame is assigned a probability score. Finally, a set of key frames is selected based on the probability scores as per user provided skimming ratio or system default skimming ratio. The proposed approach is evaluated using three benchmark datasets-BL-7F, Office, and Lobby. The experimental results show that our approach outperforms the state-of-the-art method in terms of accuracy.

A Hyperspectral (HS) image provides observational powers beyond human vision capability but represents more than 100 times the data compared to a traditional image. To transmit and store the huge volume of an HS image, we argue that a fundamental shift is required from the existing "original pixel intensity"-based coding approaches using traditional image coders (e.g., JPEG2000) to the "residual"-based approaches using a video coder for better compression performance. A modified video coder is required to exploit spatial-spectral redundancy using pixel-level reflectance modelling due to the different characteristics of HS images in their spectral and shape domain of panchromatic imagery compared to traditional videos. In this paper a novel coding framework using Reflectance Prediction Modelling (RPM) in the latest video coding standard High Efficiency Video Coding (HEVC) for HS images is proposed. An HS image presents a wealth of data where every pixel is considered a vector for different spectral bands. By quantitative comparison and analysis of pixel vector distribution along spectral bands, we conclude that modelling can predict the distribution and correlation of the pixel vectors for different bands. To exploit distribution of the known pixel vector, we estimate a predicted current spectral band from the previous bands using Gaussian mixture-based modelling. The predicted band is used as the additional reference band together with the immediate previous band when we apply the HEVC. Every spectral band of an HS image is treated like it is an individual frame of a video. In this paper, we compare the proposed method with mainstream encoders. The experimental results are fully justified by three types of HS dataset with different wavelength ranges. The proposed method outperforms the existing mainstream HS encoders in terms of rate-distortion performance of HS image compression.

The emerging High Efficiency Video Coding (HEVC) standard introduces a number of innovative and powerful coding tools to acquire better compression efficiency compared to its predecessor H.264. The encoding time complexities have also increased multiple times that is not suitable for realtime video coding applications. To address this limitation, this paper employs a novel coding strategy to reduce the time complexity in HEVC encoder by efficient selection of appropriate block-partitioning modes based on human visual features (HVF). The HVF in the proposed technique comprise with human visual attention modelling-based saliency feature and phase correlation-based motion features. The features are innovatively combined through a fusion process by developing a content-based adaptive weighted cost function to determine the region with dominated motion/saliency (RDMS)- based binary pattern for the current block. The generated binary pattern is then compared with a codebook of predefined binary pattern templates aligned to the HEVC recommended block-paritioning to estimate a subset of inter-prediction modes. Without exhaustive exploration of all modes available in the HEVC standard, only the selected subset of modes are motion estimated and motion compensated for a particular coding unit. The experimental evaluation reveals that the proposed technique notably down-scales the average computational time of the latest HEVC reference encoder by 34% while providing similar rate-distortion (RD) performance for a wide range of video sequences.