Video Images Research Articles

Analysis of Convolutional Neural Network-based Image Classifications: A Multi-Featured Application for Rice Leaf Disease Prediction and Recommendations for Farmers

This study presents a novel method for improving rice disease classification using 8 different convolutional neural network (CNN) algorithms, which will further the field of precision agriculture. A thorough investigation of deep learning methods is carried out using the UCI dataset in order to create a reliable and effective model that can correctly identify a range of rice diseases. The suggested transfer learning models performs better at identifying subtle features and complex patterns in the dataset, which results in extremely accurate disease classification. Moreover, the study goes beyond the creation of models by incorporating an intuitive Tkinter-based application that offers farmers a feature-rich interface. With the help of this cutting-edge application, farmers will be able to make timely and well-informed decisions by enabling real-time disease prediction and providing personalized recommendations. Together with the user-friendly Tkinter interface, the smooth integration of cutting-edge CNN transfer Learning algorithms-based technology that include ResNet-50, InceptionV3, VGG16, MobileNetv2 with the UCI dataset represents a major advancement toward modernizing agricultural practices and guaranteeing sustainable crop management. Remarkable outcomes include 75% accuracy for ResNet-50, 90% accuracy for DenseNet121, 84% accuracy for VGG16, 95.83% accuracy for MobileNetV2, 91.61% accuracy for DenseNet169, and 86% accuracy for InceptionV3. These results give a concise summary of the models' capabilities, assisting researchers in choosing appropriate strategies for precise and successful rice crop disease identification. A severe overfitting has been seen on VGG19 with 70% accuracy and Nasnet with 80.02% accuracy. On Renset101 only an accuracy of 54% could be achieved along with only 33% on efficientNetB0. MobileNetV2 trained model was successfully deployed on a tkinter GUI application to make predictions using image or real time video capturing.

Application of Video Abnormal Behavior Detection Algorithm in Evaluation of Track and Field Teaching and Training Effect

With the development of track and field, people pay more and more attention to the quality of classroom teaching of track and field technology, and the evaluation of teaching quality plays a key role in it. In today's educational reform, teaching evaluation plays an important role as an important method to test teachers' teaching and students' learning. With the rapid development of machine learning, especially deep learning, image-based individual abnormal behavior detection technology is becoming more and more mature, but there are still many difficulties to be solved in video-based group abnormal behavior detection technology. Therefore, it is necessary to study the detection algorithm. Based on machine vision theory, image processing theory and video analysis technology, this paper studies three key technologies involved in human abnormal behavior detection in video: moving target detection, moving target tracking and abnormal behavior detection.

Video Art and Collective Memory in Turkey

The aim of this article is to present and study contemporary video practices since the 2000s in Turkey through the representation of time and collective memory. The article features the four video artists who have structured the video art scene in Turkey: Ali Kazma, Halil Altındere, Erkan Özgen and Seza Paker. This research addresses two interconnected issues: firstly, the purpose is to examine how the distinct technical parameters of video production shape a synthesized temporality. This inquiry naturally gives rise to a secondary purpose: in what manner does this constructed temporality facilitate the exploration of memory and historical discourse? The hypothesis is that the video image, both as a technical artifact and a conceptual construct, inherently possesses its unique framework for temporal construction, thereby actively contributing to novel reinterpretations of collective memory and historiography The methodology involves conducting a formal analysis of the video content. This approach focuses on examining the visual elements, composition, use of color, sound, editing techniques, and overall aesthetics of the video to gain deeper insights into its artistic significance and meaning. By employing formal analysis, the aim of this article is to decipher the visual language employed by the video artists and uncover nuanced layers of interpretation.

Promote computer vision applications in pig farming scenarios: high-quality dataset, fundamental models, and comparable performance1

Computer vision is widely recognized as an influential technology in the field of precision management of animals. Emerging studies have demonstrated the potential to improve pig health and welfare through animal surveillance systems and computer vision (CV) algorithms. However, the lack of benchmark datasets and robust fundamental algorithms restrict CV applications for the commercial use. This study aims to bridge the gap between technology development and commercial applications in pig farming scenarios by introducing a general-purpose dataset (PigLife), comparing benchmark performances of foundational CV algorithms and model development workflows. The PigLife dataset contains video clips and images (38 short video clips, 2K image frames, 22K pig instances) across most pig production phases in a typical commercial pig farm: Breeding and Gestation, Farrow to Wean, Weaning & Nursery, and Growth to Finish. Three detection algorithms (Faster R-CNN, RetinaNet, TridentNet) and three segmentation algorithms (Mask R-CNN, MViTv2, Point-Rend) were trained on the PigLife dataset from scratch. Fine-tuning of pre-trained models (YOLO8-m, Faster-RCNN-r50) and no-training from zero-shot models (CLIP-SAM, Grouddino-HQSAM) were also evaluated to suggest faster CV development workflows for commercial applications in pig farming. This study emphasizes the necessity of a benchmark dataset for evaluating the robustness of algorithms and identifying the remaining difficulties and challenges across various algorithms. Furthermore, developing CV models from pre-trained algorithms or zero-shot models showed better performance and a faster process, which could reduce barriers when developing high-performance CV products in pig production industry.

SomaSeg: a robust neuron identification framework for two-photon imaging video

Objective.Accurate neuron identification is fundamental to the analysis of neuronal population dynamics and signal extraction in fluorescence videos. However, several factors such as severe imaging noise, out-of-focus neuropil contamination, and adjacent neuron overlap would impair the performance of neuron identification algorithms and lead to errors in neuron shape and calcium activity extraction, or ultimately compromise the reliability of analysis conclusions.Approach.To address these challenges, we developed a novel cascade framework named SomaSeg. This framework integrates Duffing denoising and neuropil contamination defogging for video enhancement, and an overlapping instance segmentation network for stacked neurons differentiating.Main results.Compared with the state-of-the-art neuron identification methods, both simulation and actual experimental results demonstrate that SomaSeg framework is robust to noise, insensitive to out-of-focus contamination and effective in dealing with overlapping neurons in actual complex imaging scenarios.Significance.The SomaSeg framework provides a widely applicable solution for two-photon video processing, which enhances the reliability of neuron identification and exhibits value in distinguishing visually confusing neurons.

Enhanced river suspended sediment concentration identification via multimodal video image, optical flow, and water temperature data fusion

Monitoring suspended sediment concentration (SSC) in rivers is pivotal for water quality management and sustainable river ecosystem development. However, achieving continuous and precise SSC monitoring is fraught with challenges, including low automation, lengthy measurement processes, and high cost. This study proposes an innovative approach for SSC identification in rivers using multimodal data fusion. We developed a robust model by harnessing colour features from video images, motion characteristics from the Lucas–Kanade (LK) optical flow method, and temperature data. By integrating ResNet with a mixed density network (MDN), our method fused the image and optical flow fields, and temperature data to enhance accuracy and reliability. Validated at a hydropower station in the Xinjiang Uygur Autonomous Region, China, the results demonstrated that while the image field alone offers a baseline level of SSC identification, it experiences local errors under specific conditions. The incorporation of optical flow and water temperature information enhanced model robustness, particularly when coupling the image and optical flow fields, yielding a Nash–Sutcliffe efficiency (NSE) of 0.91. Further enhancement was observed with the combined use of all three data types, attaining an NSE of 0.93. This integrated approach offers a more accurate SSC identification solution, enabling non-contact, low-cost measurements, facilitating remote online monitoring, and supporting water resource management and river water–sediment element monitoring.

The Articulatory and Acoustic Representation of Second-Language French Vowels.

This study examines how L1 English-L2 French learners use L1 articulatory and acoustic categories to produce L2 vowels that are both similar to and different from their L1 vowels. Previous studies examining the relationship between L1 and L2 sound inventories have found that learners reuse L1 phone categories to produce L2 phones that are perceived as similar, but importantly, there is a lack of articulatory data included in these types of studies, which has reinforced the assumption that vowel categories can be solely represented by their acoustic properties. The present study uses ultrasound tongue imaging data and videos of lip rounding in addition to acoustic data to examine how L1 English-L2 French learners produce the French vowels /i y u e ø o/ compared with their English vowels /i u e o/. The results focus on individual paths to category formation to show how learners articulate L2 vowels, and reveal that they tend to reuse L1 tongue body gestures to produce the French vowels /i u e o/, and lip rounding gestures to produce the round vowels /y u o/. This study demonstrates that transfer of articulatory gestures depends on vowel quality and emphasizes the importance of using articulatory data to inform theories of L2 category formation.

Rate dependent C* fracture parameter using the optimized wedge-split test geometry and vision-based automated crack tip detection

Typical monotonic load-line displacement (LLD) control and crack mouth opening displacement (CMOD) control cracking tests are limited to singular loading rates and testing temperatures. These experimental protocols are not designed to derive a fundamental fracture parameter capable of relating to the viscoelastic property and cracking behavior of asphalt concrete (AC). On the other hand, a multi-rate LLD-control monotonic fracture experiment is capable of capturing the viscoelastic cracking behavior of AC by providing the rate-dependent fundamental C* fracture parameter as an output. Crack length and crack speed parameters are combined with reaction loads to calculate the rate-dependent C* parameter. The current C* experiment protocol implements manual processing of video images and exhibits high variability. The study aims to optimize the wedge-split geometry and improve the C* fracture parameter calculation protocol by applying an automated vision-based crack detection system and recording the crack growth path, considering the horizontal and vertical movement of the crack tip with higher accuracy. The three notch lengths (5, 15, and 30 mm) were evaluated. Plant and laboratory-produced mixes with varying stiffness and fracture characteristics were included in the testing program. An optimized notch length for the test geometry was determined based on the on-specimen strain field observed using the digital image correlation (DIC) technique and crack path analysis. The optimum notch length was found to be 15 mm due to consistent strain localization at the crack tip and less crack deviation in the horizontal direction. The main contribution of this paper is the proposed vision-based crack path detection method, which enables measuring the crack path's true crack length and maximum horizontal deviation. The developed protocol and optimization of the testing geometry led to a significant improvement of the existing C* testing protocol, resulting in higher data quality and consistent results. The proposed protocol was validated using three comparable and distinctively different lab-produced mixtures. The results show that the C* experiment effectively differentiated between the mixes based on crack speed and the C* parameter.

Brain Activity in Visual-Motor Illusions With Enhanced Joint Motion Intensity.

Background Visual-motor illusion (VMI) is a cognitive approach used to evoke kinesthetic sensations. Research suggests that VMI can modulate brain activity depending on the specific joint movement observed. This study aimed to identify differences in brain activity when observing video images of joint movements at different intensities of movement in VMI. Methodology The study included 14 healthy adult participants. Two types of video images were used: pure ankle dorsiflexion movements (Standard-VMI) and ankle dorsiflexion movements with added resistance (Power-VMI). The brain activity measurement protocol employed a block design with one set of 15 seconds rest, 30 seconds VMI task, and 30 seconds follow-up. Each participant performed the VMI task twice, alternating between Standard-VMI and Power-VMI. Brain activity was measured using functional near-infrared spectroscopy, focusing on motor-related regions. Subjective impressions were assessed using visual analog scales (VAS) for kinesthetic illusions. Results The results revealed that Power-VMI stimulated significantly greater brain activity in the premotor and supplementary motor cortex, supramarginal gyrus, and superior parietal lobule compared with Standard-VMI. Power-VMI resulted in higher VAS values for kinesthetic illusion than Standard-VMI. Additionally, a positive correlation was observed between brain activity in the superior parietal lobule and the degree of kinesthetic illusion. Conclusions These findings indicate that Power-VMI enhances both motor-related brain areas and motor-sensory illusions, potentially having a greater impact on improving motor function. This study provides valuable insights for developing VMI interventions for rehabilitation, particularly for individuals with paralysis or movement impairments.

Dye-less quantification of tissue perfusion by laser speckle contrast imaging is equivalent to quantified indocyanine green in a porcine model.

Subjective surgeon interpretation of near-infrared perfusion video is limited by low inter-observer agreement and poor correlation to clinical outcomes. In contrast, quantification of indocyanine green fluorescence video (Q-ICG) correlates with histologic level of perfusion as well as clinical outcomes. Measuring dye volume over time, however, has limitations, such as it is not on-demand, has poor spatial resolution, and is not easily repeatable. Laser speckle contrast imaging quantification (Q-LSCI) is a real-time, dye-free alternative, but further validation is needed. We hypothesize that Q-LSCI will distinguish ischemic tissue and correlate over a range of perfusion levels equivalent to Q-ICG. Nine sections of intestine in three swine were devascularized. Pairs of indocyanine green fluorescence imaging and laser speckle contrast imaging video were quantified within perfused, watershed, and ischemic regions. Q-ICG used normalized peak inflow slope. Q-LSCI methods were laser speckle perfusion units (LSPU), the base unit of laser speckle imaging, relative perfusion units (RPU), a previously described methodology which utilizes an internal control, and zero-lag normalized cross-correlation (X-Corr), to investigate if the signal deviations convey accurate perfusion information. We determine the ability to distinguish ischemic regions and correlation to Q-ICG over a perfusion gradient. All modalities distinguished ischemic from perfused regions of interest; Q-ICG values of 0.028 and 0.155 (p < 0.001); RPU values of 0.15 and 0.68 (p < 0.001); and X-corr values of 0.73 and 0.24 (p < 0.001). Over a range of perfusion levels, RPU had the best correlation with Q-ICG (r = 0.79, p < 0.001) compared with LSPU (r = 0.74, p < 0.001) and X-Corr (r = 0.46, p < 0.001). These results demonstrate that Q-LSCI discriminates ischemic from perfused tissue and represents similar perfusion information over a broad range of perfusion levels comparable to clinically validated Q-ICG. This suggests that Q-LSCI might offer clinically predictive real-time dye-free quantification of tissue perfusion. Further work should include validation in histologic studies and human clinical trials.

Photogrammetry of the Deep Seafloor from Archived Unmanned Submersible Exploration Dives

Large amounts of video images have been collected for decades by scientific and governmental organizations in deep (&gt;1000 m) water using manned and unmanned submersibles and towed cameras. The collected images were analyzed individually or were mosaiced in small areas with great effort. Here, we provide a workflow for utilizing modern photogrammetry to construct virtual geological outcrops hundreds or thousands of meters in length from these archived video images. The photogrammetry further allows quantitative measurements of these outcrops, which were previously unavailable. Although photogrammetry had been carried out in recent years in the deep sea, it had been limited to small areas with pre-defined overlapping dive paths. Here, we propose a workflow for constructing virtual outcrops from archived exploration dives, which addresses the complicating factors posed by single non-linear and variable-speed vehicle paths. These factors include poor navigation, variable lighting, differential color attenuation due to variable distance from the seafloor, and variable camera orientation with respect to the vehicle. In particular, the lack of accurate navigation necessitates reliance on image quality and the establishment of pseudo-ground-control points to build the photogrammetry model. Our workflow offers an inexpensive method for analyzing deep-sea geological environments from existing video images, particularly when coupled with rock samples.

SE-CBAM-YOLOv7: An Improved Lightweight Attention Mechanism-Based YOLOv7 for Real-Time Detection of Small Aircraft Targets in Microsatellite Remote Sensing Imaging

Addressing real-time aircraft target detection in microsatellite-based visible light remote sensing video imaging requires considering the limitations of imaging payload resolution, complex ground backgrounds, and the relative positional changes between the platform and aircraft. These factors lead to multi-scale variations in aircraft targets, making high-precision real-time detection of small targets in complex backgrounds a significant challenge for detection algorithms. Hence, this paper introduces a real-time aircraft target detection algorithm for remote sensing imaging using an improved lightweight attention mechanism that relies on the You Only Look Once version 7 (YOLOv7) framework (SE-CBAM-YOLOv7). The proposed algorithm replaces the standard convolution (Conv) with a lightweight convolutional squeeze-and-excitation convolution (SEConv) to reduce the computational parameters and accelerate the detection process of small aircraft targets, thus enhancing real-time onboard processing capabilities. In addition, the SEConv-based spatial pyramid pooling and connected spatial pyramid convolution (SPPCSPC) module extracts image features. It improves detection accuracy while the feature fusion section integrates the convolutional block attention module (CBAM) hybrid attention network, forming the convolutional block attention module Concat (CBAMCAT) module. Furthermore, it optimizes small aircraft target features in channel and spatial dimensions, improving the model’s feature fusion capabilities. Experiments on public remote sensing datasets reveal that the proposed SE-CBAM-YOLOv7 improves detection accuracy by 0.5% and the mAP value by 1.7% compared to YOLOv7, significantly enhancing the detection capability for small-sized aircraft targets in satellite remote sensing imaging.

Differential Diagnosis of Pathological Type of Peripheral Lung Cancer with Multimodal Contrast-Enhanced Ultrasound

ObjectiveThe goal of the work described here was to investigate the role of multimodal contrast-enhanced ultrasound in the differential diagnosis of peripheral lung cancer. MethodsFrom April 2017 to July 2021, 109 patients with confirmed pulmonary malignant lesions who underwent CEUS examination were involved in our study. Seven patients were excluded because of the short duration of CEUS video or unsatisfactory imaging. Finally,102 patients with peripheral lung cancer were enrolled in this study. The maximum diameter of the lesions ranged from 1.6 to 13.0 cm (mean 6.2 ± 2.3 cm). On the basis of the pathological results, the patients were divided into the small cell lung cancer (SCLC) group and non-small cell lung cancer (NSCLC) group (including adenocarcinoma, lung squamous cell carcinoma and large cell neuroendocrine carcinoma). A Logiq E9 ultrasonic machine equipped with a 3.5 to 5.0 MHz C5-1 probe was used. Patient clinical information, CEUS features, CPI patterns and TIC parameters were analyzed and compared between different groups. Statistical analyses were performed with SPSS software and MedCalc software. The receiver operating characteristic curve was plotted. ResultsIn the differential diagnosis of SCLC and NSCLC, color parametric imaging indicated great performance. NSCLC exhibited a centripetal enhancement pattern more frequently (72.7%), while SCLC exhibited an eccentric enhancement pattern more frequently (92.9%) (p < 0.001). In the differential diagnosis of adenocarcinoma and squamous cell carcinoma, logistic regression analysis revealed that patient age of onset ≤60 y, difference in arrival time between lung and tumor ≤3.8 s, drop time of the time–intensity curve >23.2 s and absence of internal necrosis on CEUS were independent predictors for adenocarcinoma (area under the curve = 0.861). ConclusionIn our study, multimodal contrast-enhanced ultrasound provided useful information in the differential diagnosis between small cell lung cancer and non-small cell lung cancer, especially between adenocarcinoma and squamous cell carcinoma.

Modern Methods and Technical Instruments of Ecological Monitoring of the Estuaries of Small Rivers

At present, river estuaries are constantly being explored; therefore, it is necessary to receive up-todate information on their actual environmental condition, hydrological regime, and geographical features. Therefore, it is important to develop and improve the methods and technical instruments for monitoring river estuaries. The article presents a developed methodology for integrated research of small-river estuaries using new methods and techniques. The article discusses the experience in using modern domestic devices for solving practical problems of integrated water-environmental monitoring with a case study of expeditionary work in estuaries of the Chernaya River (Crimean Peninsula, Sevastopol). The possibilities of the following technical instruments used are described: The GAP-AK-12R CTD (MHI RAS), the Condor biophysical complex (Aquastandard), and the Garmin echo sounder, which make it possible to obtain field data on turbidity, the concentration of suspended particulate matter, speed and direction of currents, the average size of suspended particles, and salinity and temperature of the aquatic environment. The results of applying a new method for determining the flow velocity and water consumption based on video image processing are presented. Application of the developed methodology makes it possible to obtain the information necessary to analyze the hydrological regime and ecological state of small-river estuaries.

Application of Digital Tools and Artificial Intelligence to the Myasthenia Gravis Core Examination.

Advances in video image analysis and artificial intelligence provide the opportunity to transform the approach to patient evaluation through objective digital evaluation. We assessed ability to quantitate Zoom video recordings of a standardized neurological examination the myasthenia gravis core examination (MG-CE), which had been designed for telemedicine evaluations. We used Zoom (Zoom Video Communications) videos of patients with myasthenia gravis undergoing the MG-CE. Computer vision in combination with artificial intelligence methods were used to build algorithms to analyze videos with a focus on eye or body motions. For the assessment of examinations involving vocalization, signal processing methods were developed, including natural language processing. A series of algorithms were built that could automatically compute the metrics of the MG-CE. Fifty-one patients with MG with videos recorded twice on separate days and 15 control subjects were assessed once. We were successful in quantitating lid, eye, and arm positions and as well as well as develop respiratory metrics using breath counts. Cheek puff exercise was found to be of limited value for quantitation. Technical limitations included variations in illumination, bandwidth, and recording being done on the examiner side, not the patient. Several aspects of the MG-CE can be quantitated to produce continuous measures via standard Zoom video recordings. Further development of the technology offer the ability for trained, non-physician, health care providers to perform precise examination of patients with MG outside the clinic, including for clinical trials. Advances in video image analysis and artificial intelligence provide the opportunity to transform the approach to patient evaluation. Here, we asked whether video recordings of the typical telemedicine examination for the patient with myasthenia gravis be used to quantitate examination findings? Despite recordings not made for purpose, we were able to develop and apply computer vision and artificial intelligence to Zoom recorded videos to successfully quantitate eye muscle, facial muscle, and limb fatigue. The analysis also pointed out limitations of human assessments of bulbar and respiratory assessments. The neuromuscular examination can be enhanced by advance technologies, which have the promise to improve clinical trial outcome measures as well as standard care.

Indocyanine green fluorescence perfusion testing in robot-assisted hepatic arterial infusion pump placement

BackgroundHepatic arterial infusion pump (HAIP) treatment is a technique used to treat liver localized malignancy with intra-arterial chemotherapy. Methylene blue is generally administered to verify hepatic perfusion and exclude inadvertent extrahepatic perfusion. The use of indocyanine green dye (ICG) combined with near-infrared (NIR) fluorescence imaging during robot-assisted HAIP placement may be an attractive alternative by providing high contrast without blue discoloration of the operative field.MethodsData was collected retrospectively from 2 centers in the Netherlands. Intraoperative perfusion of the liver segments and extrahepatic perfusion were assessed using ICG/NIR as well as methylene blue on video imaging and correlated to postoperative 99 m-Tc perfusion scintigraphy.Results13 patients underwent robot-assisted surgery for HAIP placement; median length of stay was 4 days, complications occurred in 4 patients. Hepatic perfusion showed identical patterns when ICG was compared with methylene blue. In 1 patient, additional extrahepatic perfusion was found using ICG, leading to further vessel ligation. Intraoperative ICG perfusion was concordant with 99 m-Tc perfusion scintigraphy.DiscussionLiver and extrahepatic perfusion determined by ICG fluorescence imaging is concordant with blue dye perfusion and 99 m-Tc perfusion scintigraphy. Therefore, ICG fluorescence imaging is deemed a safe and reliable technique for perfusion testing during robot-assisted HAIP placement.Graphical abstract

TIHN: Tensor Improved Huber Norm for low-rank tensor recovery

Tensor Robust Principal Component Analysis (TRPCA) has received much attention in many real-world applications which aims to recover a low-rank tensor corrupted by sparse noise. Most existing TRPCA methods usually regularize the low-rank component by minimizing its Tensor Nuclear Norm (TNN). However, the original TNN shrinks all of the singular values with the same penalty which generally leads to suboptimal results. The reason is that the large singular values should be less penalized since they usually correspond to salient information of the real-world tensor. In this work, we develop a Tensor Improved Huber Norm (TIHN) which considers the difference between the singular values of tensor. The TIHN can achieve better performance by recovering the large and significant singular values exactly. We also establish the Huber TRPCA (HTRPCA) method by utilizing the proposed TIHN. In addition, an efficient optimization algorithm based on the half-quadratic theory and Alternating Direction Method of Multipliers (ADMM) framework is designed to implement the HTRPCA. Finally, experiments on color image recovery and video recovery validate the effectiveness of the proposed method.

Deep learning pose detection model for sow locomotion

Lameness affects animal mobility, causing pain and discomfort. Lameness in early stages often goes undetected due to a lack of observation, precision, and reliability. Automated and non-invasive systems offer precision and detection ease and may improve animal welfare. This study was conducted to create a repository of images and videos of sows with different locomotion scores. Our goal is to develop a computer vision model for automatically identifying specific points on the sow's body. The automatic identification and ability to track specific body areas, will allow us to conduct kinematic studies with the aim of facilitating the detection of lameness using deep learning. The video database was collected on a pig farm with a scenario built to allow filming of sows in locomotion with different lameness scores. Two stereo cameras were used to record 2D videos images. Thirteen locomotion experts assessed the videos using the Locomotion Score System developed by Zinpro Corporation. From this annotated repository, computational models were trained and tested using the open-source deep learning-based animal pose tracking framework SLEAP (Social LEAP Estimates Animal Poses). The top-performing models were constructed using the LEAP architecture to accurately track 6 (lateral view) and 10 (dorsal view) skeleton keypoints. The architecture achieved average precisions values of 0.90 and 0.72, average distances of 6.83 and 11.37 in pixel, and similarities of 0.94 and 0.86 for the lateral and dorsal views, respectively. These computational models are proposed as a Precision Livestock Farming tool and method for identifying and estimating postures in pigs automatically and objectively. The 2D video image repository with different pig locomotion scores can be used as a tool for teaching and research. Based on our skeleton keypoint classification results, an automatic system could be developed. This could contribute to the objective assessment of locomotion scores in sows, improving their welfare.

High-performance real-time fire detection and forecasting framework for industrial cables

In industrial scenarios, cable fires have always been the most common threat, and traditional fire detection systems often rely on a large number of sensors, and the detection range is very limited, and it is impossible to effectively predict the fire situation in time. In this paper, we propose a detection and prediction scheme for industrial cable fire, which breaks the limitations of the previous research on multi-sensor signal input, and highly couples the detection and prediction modules to realize fire prediction based on video image input only. In fire detection, we design an object detection model using HSV for flame feature enhancement based on YOLOv8, and in fire prediction aspect, we use iTransformer as a time series prediction model to mine the correlation between various parameters to predict the spread of fire. In the experiments, the average absolute percentage error of the flame detection model for the detection of flame height, width and longitudinal position was 3.49%–10.64 %, 2.45%–8.89 % and 1.61%–9.31 %, respectively, and the MAPE of the time series prediction model for the above three parameters was 11.18%–15.06 % and 4.35%–8.18 %, 3.37%–6.62 %.The results of the above experiments verify that the proposed model has the ability to quantitatively analyze the fire spread trend in the actual fire and help firefighters make decisions.

3D-Oculography: New Method of Determination of Human Gaze Point Position in Space

INTRODUCTION: Oculography — a method of recording movement of eyeballs of a human or animal by analyzing changes in electrical potentials recorded by two electrodes fixed on skin near the eye socket is used to solve various problems. Among them are the determination of pronounced nystagmus, of characteristic changes of oculomotor reactions in different conditions of observing video images. Of interest is the determination of the position of gaze point (GP) and of the area of increased attention in three-dimensional space. This information is associated with the cognitive system of the observer and is of interest not only to physiologists, but also to specialists of related fields. . AIM: To experimentally prove the effectiveness of the developed method of 3D-oculography providing determination of the GP position in space. MATERIALS AND METHODS: The possibility of restoring GP position based on the analysis of recorded oculograms even in the presence of additive noise induced by the external electromagnetic fields, is proven by the numeric modeling method. Direct experimental studies were conducted with recording oculograms of a human observer on a previously verified equipment. A ZB-2 oculograph was used, the data obtained were verified by the method of average values. RESULTS: The obtained modeling results permitted to determine the maximal values of dispersion of additive noise and amplitude shifts of signals, which make possible satisfactory restoration of the coordinates of the GP and parameters of its movement trajectory. The qualitative correspondence of the experimental results to the results of numerical modeling was proven. The correspondence between trajectories of the observed object moving in space and trajectories of GP movement synthesized from counts of recorded oculograms was confirmed. CONCLUSION: Multichannel record of oculographic signals permits restoring parameters of the GP trajectory. Creating systems for recording these signals requires minimization of the noise level. An increase in dispersion of the noise component of the signal leads to most significant errors in calculating the coordinates of trajectory points.