Computer vision based train rolling stock examination

We propose a variational framework for the integration of multiple competing shape priors into level set based segmentation schemes. By optimizing an appropriate cost functional with respect to both a level set function and a (vector-valued) labeling function, we jointly generate a segmentation (by the level set function) and a recognition-driven partition of the image domain (by the labeling function) which indicates where to enforce certain shape priors. Our framework fundamentally extends previous work on shape priors in level set segmentation by directly addressing the central question of where to apply which prior. It allows for the seamless integration of numerous shape priors such that--while segmenting both multiple known and unknown objects--the level set process may selectively use specific shape knowledge for simultaneously enhancing segmentation and recognizing shape.

Tracking of regions and object boundaries in an image sequence is a well studied problem in image processing and computer vision. So far, numerous approaches tracking different features of the objects (contours, regions or points of interest) have been presented. Most of these approaches have problems with robustness. Typical reasons are noisy images, objects with identical features or partial occlusions of the tracked features. In this paper we propose a novel level set based tracking approach, that allows robust tracking on noisy images. Our framework is able to track multiple regions in an image sequence, where a level set function is assigned to every region. For already known or learned objects, transformation invariant shape priors can be added to ensure a robust tracking even under partial occlusions. Furthermore, we introduce a simple decision function to maintain the desired topology for multiple regions. Experimental results demonstrate the method for arbitrary numbers of shape priors. The approach can even handle full occlusions and objects which are temporarily hidden in containers.KeywordsActive ContourNoisy ImageRobust TrackingShape PriorGeodesic Active ContourThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Purpose. To analyze the results of the study of dynamic parameters of a gondola car, model 12-1704-04 with axle load 23.5 ton in bogies, models 18-1750 and two gondola cars, model 12-1905 with axial load of 25 ton, one of which is equipped with bogies, model 18-1711 with friction wedges of spring suspension with spatial form with increased angle to the horizontal line of intersection of the contact surfaces between the wedge and bolster, the other gondola car, model 12-1905 is equipped with bogies of the same model on which the friction wedges fitted with a flat form of contact surface. It has an angle of inclination like a wedge of bogie, model 18-100. On the basis of the obtained results to draw conclusions about the feasibility of unification design bogie bolster, model 18-1711 with bogie, model 18-100 by contact surfaces with elements of spring suspension. Methodology. Research on dynamic performance of cars was performed during running dynamic tests of specimens of freight cars in experimental train consisting of two locomotives, a laboratory, and three gondola cars of the above mentioned models. Findings. Main results of dynamic studies are presented as graphs of indicators on the speed of the train and the experimental evidence that the freight gondola cars on bogies, model 18-1711 with flat-shaped wedges, in most cases are better than the others. Originality. Research results of cars on bogies, model 18-1711 were obtained. They let assess the dependence of the dynamic performance of the car from the design of the friction wedges of spring suspension. Practical value. Cast parts of bogie, model 18-1711 with 25 ton axle load can be used as a replacement of defective parts of bogie, model 18-100 and its analogs.

The automatic segmentation of medical images has widespread applications in modern clinical workflows. The Segment Anything Model (SAM), a recent development of foundational models in computer vision, has become a universal tool for image segmentation without the need for specific domain training. However, SAM's reliance on prompts necessitates human-computer interaction during the inference process. Its performance on specific domains can also be limited without additional adaptation. In contrast, traditional models like nnUNet are designed to perform segmentation tasks automatically during inference and can work well for each specific domain, but they require extensive training on domain-specificdatasets. To leverage the advantages of both foundational and domain-specific models and achieve fully automated segmentation with limited training samples, we propose nnSAM, which combines the robust feature extraction capabilities of SAM with the automatic configuration abilities of nnUNet to enhance the accuracy and robustness of medical image segmentation on smalldatasets. We propose the nnSAM model for small sample medical image segmentation. We made optimizations for this goal via two main approaches: first, we integrated the feature extraction capabilities of SAM with the automatic configuration advantages of nnUNet, which enables robust feature extraction and domain-specific adaptation on small datasets. Second, during the training process, we designed a boundary shape supervision loss based on level set functions and curvature calculations, enabling the model to learn anatomical shape priors from limited annotationdata. We conducted quantitative and qualitative assessments on the performance of our proposed method on four segmentation tasks: brain white matter, liver, lung, and heart segmentation. Our method achieved the best performance across all tasks. Specifically, in brain white matter segmentation using 20 training samples, nnSAM achieved the highest DICE score of 82.77 ( 10.12) % and the lowest average surface distance (ASD) of 1.14 ( 1.03) mm, compared to nnUNet, which had a DICE score of 79.25 ( 17.24) % and an ASD of 1.36 ( 1.63) mm. A sample size study shows that the advantage of nnSAM becomes more prominent under fewer trainingsamples. A comprehensive evaluation of multiple small-sample segmentation tasks demonstrates significant improvements in segmentation performance by nnSAM, highlighting the vast potential of small-samplelearning.

Level set methods have been widely used in image processing and computer vision. In conventional level set formulations, the level set function typically develops irregularities during its evolution, which may cause numerical errors and eventually destroy the stability of the evolution. Therefore, a numerical remedy, called reinitialization, is typically applied to periodically replace the degraded level set function with a signed distance function. However, the practice of reinitialization not only raises serious problems as when and how it should be performed, but also affects numerical accuracy in an undesirable way. This paper proposes a new variational level set formulation in which the regularity of the level set function is intrinsically maintained during the level set evolution. The level set evolution is derived as the gradient flow that minimizes an energy functional with a distance regularization term and an external energy that drives the motion of the zero level set toward desired locations. The distance regularization term is defined with a potential function such that the derived level set evolution has a unique forward-and-backward (FAB) diffusion effect, which is able to maintain a desired shape of the level set function, particularly a signed distance profile near the zero level set. This yields a new type of level set evolution called distance regularized level set evolution (DRLSE). The distance regularization effect eliminates the need for reinitialization and thereby avoids its induced numerical errors. In contrast to complicated implementations of conventional level set formulations, a simpler and more efficient finite difference scheme can be used to implement the DRLSE formulation. DRLSE also allows the use of more general and efficient initialization of the level set function. In its numerical implementation, relatively large time steps can be used in the finite difference scheme to reduce the number of iterations, while ensuring sufficient numerical accuracy. To demonstrate the effectiveness of the DRLSE formulation, we apply it to an edge-based active contour model for image segmentation, and provide a simple narrowband implementation to greatly reduce computational cost.

Functional near-infrared spectroscopy (fNIRS) is a preferred neuroimaging technique for studies requiring high ecological validity, allowing participants greater freedom of movement. Despite its relative robustness against motion artifacts (MAs) compared to traditional neuroimaging methods, fNIRS still faces challenges in managing and correcting these artifacts. Many existing MA correction algorithms notably lack validation on real data with ground-truth movement information. In this work, we combine computer vision, ground-truth movement data, and fNIRS signals to preliminarily characterize the association between specific head movements and MAs. Fifteen participants (age = 22 . 27 ± 2 . 62 years) took part in a whole-head fNIRS study, performing controlled head movements along three main rotational axes. Movements were categorized by axis (vertical, frontal, sagittal), speed (fast, slow), and type (half, full, repeated rotation). Experimental sessions were video recorded and analyzed frame-by-frame using the SynergyNet deep neural network to compute head orientation angles. Maximal movement amplitude and speed were extracted from head orientation data, while spikes and baseline shifts were identified in the fNIRS signals. Results showed that head orientation and movement metrics extracted via computer vision closely aligned with participant instructions. Additionally, repeated as well as upward and downward movements tended to compromise fNIRS signal quality. The occipital and pre-occipital regions were particularly susceptible to MAs following upwards or downwards movements, whereas the temporal regions were most affected by bend left, bend right, left, and right movements. These findings underscore the importance of cap adherence and fit in the relationship between movements and MAs. Overall, the present work lays the foundation for an automated approach to developing and validating fNIRS MA correction algorithms.

Traffic Hazards is one of the major problems facing across the world. One of the major causes of traffic hazards is increase in vehicles and dense population. Reducing traffic hazards is one the major challenges as majority of the deaths across the world are due to road accidents. Hence there is a need to provide better transportation facilities that can reduce the ratio of road accidents and save life's of people. One of the solution that is proposed in this paper is using IR sensors and Arduino Uno technology. The system has two phases-Accident Detection and Accident Prevention. The detection phase is carried out using IR sensors that could detect and alert the people by sending SMS using GSM module that contains predefined numbers and accident location using GPS module. Second Phase, Accident prevention is carried out using IR sensors by warning the driver about the neighboring vehicles when the distance between them is beyond the threshold value. Simulation results and Prototype is presented in this paper.

Prior sleep timing and visual recognition of emotional faces in 6-month-old infants

Deep learning is a machine learning technique that allows the computer to process things that occur naturally to humans. Today, deep learning techniques are commonly used in computer vision to classify images and videos. As a result, for challenging computer vision problems, deep learning provides state of the art solutions to it. Coral reefs are an essential resource of the earth. A new study finds the planet has lost half of its coral reefs since 1950. It is necessary to restore and prevent damage to coral reefs as they play an important role in maintaining a balance in the marine ecosystem. This proposed work helps to prevent the corals from bleaching and restore them to a healthy condition by identifying the root cause of the threats. In the proposed work, using deep learning CNN techniques, the images are classified into Healthy and Stressed coral reefs. Stressed coral reefs are an intermediate state of coral reef between healthy and bleached coral reefs. The pre-trained models Resnet50 and Inception V3 are used in this study to classify the images. Also, a proposed CNN model is built and tested for the same. The results of Inception V3 and Resnet50 are improved to 70% and 55% by tuning the hypermeters such as dropouts and batch normalisation. Similarly, the proposed model is tuned as required and obtains a maximum of up to 90% accuracy. With large datasets, the optimum amount of neural networks and tuning it as required brings higher accuracy than other methods.

The simultaneous recovery of three-dimensional (3D) structure from motion (SfM) for the sequences of images, is one of the more difficult problems in computer vision. Classical approaches to the problem rely on using algebraic techniques to solve for these unknowns given two or more image. Motion analysis and 3D shape estimation based on the estimated motion is an important problem in computer vision. The correspondence problem is an important tool in SfM where in this paper a general 3D motion based on a simple rotation, tilt, roll and translation is proposed, and then is used for 3D shape estimation. The current work expands the 2D motion estimation (shift, rotation) to accommodate general 3D motion (shift, rotation, tilt and roll). The proposed work in this paper of motion estimation of moving object is based on Fourier Descriptor Transformation (FDT) analysis before and after motion. The FDT is also used to resolve the correspondence problem in a sequence of images. We test our method on several large-scale photo collections, show the efficacy of the introduced approach to improve reconstruction accuracy.

To date, multimodal speech recognition systems based on the processing of audio and video signals show significantly better results than their unimodal counterparts. In general, researchers divide the solution of the audio–visual speech recognition problem into two parts. First, in extracting the most informative features from each modality and second, in the most successful way of fusion both modalities. Ultimately, this leads to an improvement in the accuracy of speech recognition. Almost all modern studies use this approach with video data of a standard recording speed of 25 frames per second. The choice of such a recording speed is easily explained, since the vast majority of existing audio–visual databases are recorded with this rate. However, it should be noticed that the number of 25 frames per second is a world standard for many areas and has never been specifically calculated for speech recognition tasks. The main purpose of this study is to investigate the effect brought by the high-speed video data (up to 200 frames per second) on the speech recognition accuracy. And also to find out whether the use of a high-speed video camera makes the speech recognition systems more robust to acoustical noise. To this end, we recorded a database of audio–visual Russian speech with high-speed video recordings, which consists of records of 20 speakers, each of them pronouncing 200 phrases of continuous Russian speech. Experiments performed on this database showed an improvement in the absolute speech recognition rate up to 3.10%. We also proved that the use of the high-speed camera with 200 fps allows achieving better recognition results under different acoustically noisy conditions (signal-to-noise ratio varied between 40 and 0 dB) with different types of noise (e.g. white noise, babble noise).

Road accidents have become an issue of major concern to the people. This paper presents an accident prevention mechanism developed through alcohol detection using an MQ3 alcohol sensor followed by automatic engine locking. The detection part uses an SW-420 vibration sensor to detect any sort of abnormal vibrations that may occur from a collision. This is accompanied by supervised deep learning CNN algorithms. The accident scene image is captured using a front camera built in the vehicle to be used by the deep learning model for accident prediction. Accident detection is followed by communication to the nearest emergency center using GPS and GSM modules.

Commercial Unmanned aerial vehicle (UAV) industry, which is publicly known as drone, has seen a tremendous increase in last few years, making these devices highly accessible to public. This phenomenon has immediately raised security concerns due to fact that these devices can intentionally or unintentionally cause serious hazards. In order to protect critical locations, the academia and industry have proposed several solutions in recent years. Computer vision is extensively used to detect drones autonomously compared to other proposed solutions such as RADAR, acoustics and RF signal analysis thanks to its robustness. Among these computer vision-based approaches, we see the preference of deep learning algorithms thanks to their effectiveness. In this paper, we are presenting an autonomous drone detection and tracking system which uses a static wide-angle camera and a lower-angle camera mounted on a rotating turret. In order to use memory and time efficiently, we propose a combined multi-frame deep learning detection technique, where the frame coming from the zoomed camera on the turret is overlaid on the wide-angle static camera’s frame. With this approach, we are able to build an efficient pipeline where the initial detection of small sized aerial intruders on the main image plane and their detection on the zoomed image plane is performed simultaneously, minimizing the cost of resource exhaustive detection algorithm. In addition to this, we present the integral system including tracking algorithms, deep learning classification architectures and the protocols.

The number of automobiles has also grown quickly to meet the enormous population. Additionally, this resulted in an increase in accidents. The accident prevention strategies now in use are all static and dated. Additionally, there is no reliable accident detection system. Timely medical aid can help in saving lives. Drive Sentinel is an advanced vehicle safety and monitoring system aimed at reducing road accidents and improving driver accountability. The system integrates driver drowsiness detection, real-time GPS tracking, and GSM-based communication to create a comprehensive in-vehicle safety solution. Using computer vision and embedded sensors, Drive Sentinel monitors the driver’s eye movements and facial cues to detect signs of fatigue or inattention. Upon detection of drowsiness or a potential accident, the system triggers immediate visual and audio alerts to wake the driver and simultaneously sends an SMS with the vehicle’s GPS location to registered emergency contacts using a GSM module. The hardware implementation involves a Raspberry Pi or microcontroller, camera module, GPS receiver, GSM modem, and buzzer, all working in coordination to ensure real-time monitoring and response. The use of machine learning for face and eye detection enhances accuracy in identifying driver fatigue. This system can be particularly beneficial for long-haul drivers, public transport operators, and fleet management companies. By combining embedded systems, IoT, and artificial intelligence, Drive Sentinel contributes to building safer roads and reducing the number of accidents caused by human error. Keywords- Drive Drowsiness Detection, Accident Prevention, Vehicle Safety System, GPS Tracking, GSM Alert System, Real-Time Monitoring, Image Processing, IoT in Transportation

Biometrics is the process of making use of physical, biological and behavioral traits for identification purposes. Biometrics finds applications in wide variety of fields such as security, access control, and military. Among the various kinds of biometrics, face biometrics is one among the fast growing biometrics due to its flexibility in usage and easy availability of equipments. This work is focused on the area of face biometrics, which incorporates the concept of image processing along with it. The proposed work is aimed to determine the recognition accuracy of face recognition technique and compare its accuracy with existing techniques and also addresses the problems related to facial details, illumination and facial expressions in face recognition. This work shows the higher and gradual increase in accuracy of face recognition and enables to implement wide variety of applications in computer vision using standard face databases.