Computer Vision Research Articles

A novel hybrid-view technique for accurate mass estimation of kimchi cabbage using computer vision

This study addresses the accurate estimation of kimchi cabbage mass, as cabbage leaves exhibit size variability and complex leaf structures. Conventional mass estimation methods, which rely solely on external imaging, often overlook leaf gaps. To improve accuracy, we propose an innovative computer vision system utilizing hybrid-view images and detailed saturation analysis. Our system quantifies the impact of leaf gaps on mass using features from the saturation channel of images of bisected cabbage. Our proposed method can be easily integrated into existing workflows and has the potential to improve labor efficiency. Our approach outperforms the conventional method (R2 of 0.66 and relative error of 8.68%), achieving a 0.92 R2 value and lowering the relative error to 4.22%. This advancement offers a robust solution for the mass estimation of kimchi cabbage and suggests potential applications for other foods and crops with internal voids.

Automatic Diagnosis of Parkinson’s Disease using Handwriting Patterns

Parkinson's Disease (PD) is a neuro-degenerative syndrome characterized by motor and non-motor signs, and early detection is crucial for effective intervention. This paper presents a novel approach for PD detection using computer vision and machine learning techniques applied to Spiral-Wave handwriting analysis. The dataset comprises frontal handwritten images obtained through the Spiral-Wave test, capturing subtle motor control differences. Our methodology involves resizing images to a standardized 200x200 pixels, converting them to grayscale, and applying thresholding for improved feature abstraction. Histogram of Oriented Gradients (HOG) is employed to capture shape and texture information. The development of a strong approach for deriving significant features from Spiral-Wave handwriting patterns and the usage of machine learning classifiers for precise PD analysis are the two main goals of this work. The emphasis is on using Random Forest and K-Nearest Neighbours (KNN) classifiers for Spiral and Wave pictures, respectively, in conjunction with the Histogram of Oriented Gradients (HOG) approach for feature extraction. For Spiral images, a Random Forest Classifier is utilized, achieving an accuracy of 86.67%. The classifier's interpretability is enhanced through an analysis of feature importance, revealing critical HOG features for distinguishing between healthy and PD-afflicted patterns. The Wave images are classified using a K-Nearest Neighbours (KNN) model, attaining an accuracy of 76.67%. Performance metrics, including precision, recall, and F1-score, offer a nuanced assessment of the KNN model's capabilities.

AR based Indoor Navigation System

Indoor navigation within college campuses poses unique challenges due to complex building layouts and varying levels of accessibility. This paper presents a student project aimed at developing an Augmented Reality based indoor navigation system (AR-INS) specifically tailored for college campuses. The system integrates computer vision, depth perception and Google’s ARCore techniques to provide users with real-time navigation guidance using their smartphones or AR-enabled devices. We describe the system architecture, implementation details, and evaluation results from user testing conducted on our college campus. The project highlights the feasibility of utilizing AR technology for enhancing campus navigation experiences and provides valuable insights for future research and development in this domain. This project addresses the growing need for efficient indoor navigation, which can significantly improve the overall user experience and accessibility in indoor spaces. The AR-INS is not only user- friendly but also adaptable to various environments, making it a valuable addition to smart building infrastructure and enhancing the quality of life for users. The implementation of this system can have an impact on how people navigate complex indoor spaces, and give an solution for complex indoor spaces. Keywords—AR-INS, computer vision, depth perception Google’s ARCore

Intrinsic-style distribution matching for arbitrary style transfer

Although arbitrary style transfer has been a hot topic in computer vision, most existing methods that directly align style and content features frequently result in unnatural effects in the generated images, such as over-stylization and style leakage. In this paper, we introduce a novel progressive Intrinsic-style Distribution Matching (ISDM) approach which initially aligns the intrinsic style distribution of both style and content images and then integrates it with the image’s content component. This novel approach can effectively alleviate the issue of over-stylization while preserving the content structure, particularly in the case of high contrast and vivid colors. To further enhance the performance of our method, we propose a learnable multi-level style modulation module to assist the network in aligning the intrinsic style distributions. Moreover, two contrastive objectives are proposed to improve the ability of the encoder to extract more distinct and representative intrinsic content and intrinsic style features. Extensive experimental results showcase that our approach can preserve more content details than state-of-the-art methods. It can also generate more natural images, especially when the style image has high contrast and vivid colors.

Unsafe-Net: YOLO v4 and ConvLSTM based computer vision system for real-time detection of unsafe behaviours in workplace

AbstractUnsafe behaviour is a leading cause of death or injury in the workplace, including many accidents. Despite regular safety inspections in workplaces, many accidents occur as a result of breaches of occupational health and safety protocols. In these environments, despite efforts to prevent accidents and losses in hazardous environments, human error cannot be completely eliminated. In particular, in computer-based solutions, automated behaviour detection has low accuracy, is very costly, not real-time and requires a lot of time. In this study, we propose Unsafe-Net, a hybrid computer vision approach using deep learning models for real-time classification of unsafe behaviours in workplace. For the Unsafe-Net, a dataset is first specifically created by capturing 39 days of video footage from a factory. Using this dataset, YOLO v4 and ConvLSTM methods are combined for object detection and video understanding to achieve fast and accurate results. In the experimental studies, the classification accuracy of unsafe behaviours using the proposed Unsafe-Net method is 95.81% and the average time for action recognition from videos is 0.14 s. In addition, the Unsafe-Net has increased the real-time detection speed by reducing the average video duration to 1.87 s. In addition, the system is installed in a real-time working environment in the factory and employees are immediately alerted by the system, both audibly and visually, when unsafe behaviour occurs. As a result of the installation of the system in the factory environment, it has been determined that the recurrence rate of unsafe behaviour has been reduced by approximately 75%.

Federated deep reinforcement learning for mobile robot navigation

Navigation of a mobile robot in an unknown environment ensuring the safety of the robot and its surroundings is of utmost importance. Traditional methods, such as pathplanning algorithms, simultaneous localization and mapping, computer vision, and fuzzy techniques, have been employed to address this challenge. However, to achieve better generalization and self-improvement capabilities, reinforcement learning has gained significant attention. The concern of privacy issues in sharing data is also rising in various domains. In this study, a deep reinforcement learning strategy is applied to the mobile robot to move from its initial position to a destination. Specifically, the Deep Q-Learning algorithm has been used for this purpose. This strategy is trained using a federated learning approach to overcome privacy issues and to set a foundation for further analysis of distributed learning. The application scenario considered in this work involves the navigation of a mobile robot to a charging point within a greenhouse environment. The results obtained indicate that both the traditional deep reinforcement learning and federated deep reinforcement learning frameworks are providing 100% success rate. However federated deep reinforcement learning could be a better alternate since it overcomes the privacy issue along with other advantages discussed in this paper.

Measurement Of Spatial-Temporal Waves In The Laboratory Using Computer Vision Technolog

Given the complexity of ocean wave phenomena, physical modeling continues to be essential in various engineering and research applications. Wave height, a characteristic measured in nearly all projects, is commonly assessed using resistance/capacitance wave gauges. However, a significant limitation of these traditional wave gauges is their point-based nature, which restricts their ability to describe detailed spatial wave characteristics, especially during wave nonlinearity or when waveforms change. Furthermore, to reconstruct the 3D wave field, an array of wave gauges is often employed alongside an appropriate interpolation algorithm. Additionally, wave gauges may alter flow/wave conditions in specific scenarios. To address these limitations, various non-intrusive measurement techniques have been proposed. Gomit et al. (2022) classified these into stereoscopic, projection, and light-based methods, including LiDAR (Blenkinsopp et al., 2012), RGB-D cameras (Martínez-Aranda et al., 2018), and binocular stereo vision (Li et al., 2022). However, their applicability is constrained by factors such as optical setup complexity and device cost. With the rapid advancement of computer vision technology, it is possible to improve these issues and better describe temporal and spatial wave field variations using such techniques. This conference contribution details the reconstruction of the temporal and spatial 3D free water surface using photogrammetry techniques applied to images from consumer webcams. It also compares these results with measurements obtained from a wave gauge.

Hybrid Transfer Learning and Support Vector Machine Models for Asphalt Pavement Distress Classification

Pavement condition evaluation plays a crucial role in assisting with the management of the highway infrastructure. However, the current methods used for assessing pavement conditions are costly, time-consuming, and subjective. There is a growing need to automate these assessment tactics and leverage low-cost technologies to enable widespread deployment. This study aims to develop robust and highly accurate models for classifying asphalt pavement distresses using transfer learning (TL) techniques based on pretrained deep learning (DL) networks. This topic has gained considerable attention in the field since 2015 when DL became the mainstream choice for various computer vision tasks. While progress has been made in TL model development, challenges persist in areas of accuracy, repeatability, and training cost. To tackle these challenges, this study proposes hybrid models that combine DL networks with support vector machines (SVMs). Three strategies were evaluated: single DL models using transfer learning (TLDL), hybrid models combining DL and SVM (DL+SVM), and hybrid models combining TLDL and SVM (TLDL+SVM). The performance of each strategy was assessed using statistical metrics based on the confusion matrix. Results consistently showed that the TLDL+SVM strategy outperformed the other approaches in accuracy and F1 scores, regardless of the DL network type. On average, the hybrid models achieved an accuracy of 95%, surpassing the 80% accuracy of the best single model and the 55% accuracy for DL+SVM without TL. The results clearly indicate that employing transfer-learned models as feature extractors, in combination with SVM as the classifier, consistently achieves exceptional performance.

Cascaded combination of total variation regularization and contrast limited adaptive histogram equalization based image dehazing

ABSTRACT Haze and fog can significantly reduce the visibility of distant objects by scattering light and creating blurry, washed-out images lacking in detail. Image dehazing is a technique used in computer vision to enhance the clarity of such obscured images. This study examines the effectiveness of various dehazing methods using real hazy images and synthetic images with artificially created haze. Performance metrics such as PSNR, SSIM, MSE, CII, and computation time are used to evaluate the proposed method. The evaluation is carried out on datasets like RESIDE, GMAN, O-Haze, I-Haze, NH-Haze, and Dense haze to compare the proposed method with existing models. The proposed method attains 27.46%, 20.63% and 21.09% higher PSNR and 12.36%, 23.95% and 36.12% lower Natural Image Quality Evaluator for RESIDE dataset when analyzed to the existing models, such as strategic method towards contrast enhancement by 2D histogram equalization under TV decomposition (CE-TDHE-TVD), multiple level framework basis contrast enhancement for uniform with non-uniform back ground imageries utilizing appropriate histogram equalization (CE-VHE), and Contrast enhancement with brightness preservation of low light pictures with the help of combined CLAHE and BPDHE histogram equalization (CE-CLAHE-BPDHE) respectively.

TARGET RECOGINITION AND DETECTION USING DEEP LEARING TECHNIQUE

Object detection is a fundamental computer vision function for identifying and locating things in images or videos. Unlike image classification, object detection not only classifies the items in an image but also locates them inside the picture by constructing a bounding box around each one. Defence objects include a wide range of key assets needed for military and security activities. Tanks, armoured personnel carriers (APCs), military trucks, planes, helicopters, naval boats, and unmanned ground vehicles (UGVs) are among the types of vehicles included. Detection of these vehicles is critical for tracking troop movements, analysing battlefield dynamics, and maintaining operational preparedness. The modern defence scenario demands improved surveillance and security measures to successfully counter growing threats. This work suggests using the You Only Look Once (YOLO) method for real-time identification of numerous defence items. The project's goal is to improve security by properly detecting a variety of items, including cars, individuals, equipment, and structures, with a single deep learning model. Methodologically, a broad dataset containing photos and videos of defence items in various locations is collected and preprocessed. The YOLO algorithm is then trained on this dataset, with parameters optimised for high accuracy and recall rates. Performance evaluation include thorough testing on various datasets to determine accuracy, speed, and resilience. KEYWORDS: Target recognition, Detection, YOLO algorithm, Military applications, Flying objects, Vehicles, Sensor integration, Real-time processing, Threat detection, Situational awareness.

Gait-Based Human Activity Recognition Using Efficient Sensor Fusion and Deep Learning Approach

Human activity recognition is an important area of computer vision research. Its applications include surveillance systems, patient monitoring systems, and a variety of systems that involve interactions between persons and electronic devices such as human-computer interfaces. Most of these applications require an automated recognition of high-level activities, composed of multiple simple (or atomic) actions of persons. A novel feature selection approach is then proposed in order to select a subset of discriminant features, construct an online activity recognizer with better generalization ability, and reduce the smartphone power consumption. Experimental results on a publicly available data set show that the fusion of both accelerometer and gyroscope data contributes to obtain better recognition performance than that of using single source data, and that the proposed feature selector outperforms three other comparative approaches in terms of four performance measures. Such activity profiling systems are dependent on classification algorithms which can effectively interpret body-worn sensor data and identify different activities. the key research challenges that human activity recognition shares with general pattern recognition and identify those challenges that are specific to human activity recognition. The aim of paper is to explore real life applications like contactless employee recognition system using gait analysis which uses sensor data as base to identify employees based on their gait movement. This requires understanding the dimensions of sensor data and its application exploring other potential real-life applications and optimizing the methodology are also one of the core objectives.

Intelligent Space: Enhancing Living Environment with Smart Technology (Smart Room)

In the era of smart living, the integration of advanced technologies with in indoor environments has become imperative. This abstract presents a comprehensive solution for transforming conventional rooms into intelligent spaces through the implementation of a visitor counter System, air quality monitor, and an IoT-based room automation system. The visitor counter System utilizes state-of-the-art sensors and computer vision algorithms to accurately track and count the number of individuals within the room. Simultaneously, an air quality Monitoring system is incorporated to assess and maintain optimal indoor air quality. The integration of an IoT-based room automation system further enhances the smart room's functionality. Through the use of interconnected smart devices, occupants can customize their environment based on preferences, contributing to energy efficiency and overall comfort. This smart room solution is designed with a focus on user-centricity, energy efficiency, and sustainability.

Assistive Technology for Navigation of Visually Impaired People

In this paper, an assistive navigational technology was proposed for those who are blind. People who are visually impaired are more likely to be physically inactive, move more slowly, and dread falling. As a result, the majority of them lack the confidence necessary to travel on their own in strange places. Despite the fact that some technology advancements can help them, not all of them are affordable to the common man. Through our efforts, we hope to help visually impaired persons feel secure and confident while navigating new surroundings. Assistive tech- nology for navigation of visually impaired people is an affordable and efficient computer vision based blind guidance system that helps the user detect the distance to obstacles well in advance and avoid them. The suggested system also includes an emergency distress alert feature that enables users to communicate their locations in an emergency.

Tor Trace in Images: A Novel Multi-Tab Website Fingerprinting Attack With Object Detection

Abstract Website fingerprinting (WF) attacks on Tor enable a passive adversary to predict the encrypted web browsing activity of a victim by matching the eavesdropped traffic with pretrained classifiers. Nowadays, deep learning-based methods have led to significant achievements in the single-tab Tor WF attack. However, the practical implementation of these single-tab methods is challenging due to most real-world Tor traffic involving multiple tabs. Existing single-tab methods hardly identify multi-tab WF due to the overlapping areas of the traffic trace confusing the original features. In this paper, we propose a Trace Image-based Object Detection model named TIOD as a novel multi-tab attacking model. Specifically, we model the traffic overlap in Tor as the object overlap in object detection tasks from the computer vision field. Besides, we propose a special S-matrix scheme to convert a trace into an image: (i) Retaining the original features by keeping the direction and order of the cells and (ii) Bringing cells of the same page closer together in space. We then utilize a specially designed object detection model for trace images, WF R-CNN, to extract features and identify potential destination websites within multi-tab traces. Comparative analysis with other multi-tab attacks is conducted, and the empirical results consistently underscore the superior performance of the proposed trace image model across diverse datasets. In the two-tab setting, TIOD achieves the best accuracy with more than 85% on both the first and second tabs.

A Computer Vision Framework for Structural Analysis of Hand-Drawn Engineering Sketches.

Structural engineers are often required to draw two-dimensional engineering sketches for quick structural analysis, either by hand calculation or using analysis software. However, calculation by hand is slow and error-prone, and the manual conversion of a hand-drawn sketch into a virtual model is tedious and time-consuming. This paper presents a complete and autonomous framework for converting a hand-drawn engineering sketch into an analyzed structural model using a camera and computer vision. In this framework, a computer vision object detection stage initially extracts information about the raw features in the image of the beam diagram. Next, a computer vision number-reading model transcribes any handwritten numerals appearing in the image. Then, feature association models are applied to characterize the relationships among the detected features in order to build a comprehensive structural model. Finally, the structural model generated is analyzed using OpenSees. In the system presented, the object detection model achieves a mean average precision of 99.1%, the number-reading model achieves an accuracy of 99.0%, and the models in the feature association stage achieve accuracies ranging from 95.1% to 99.5%. Overall, the tool analyzes 45.0% of images entirely correctly and the remaining 55.0% of images partially correctly. The proposed framework holds promise for other types of structural sketches, such as trusses and frames. Moreover, it can be a valuable tool for structural engineers that is capable of improving the efficiency, safety, and sustainability of future construction projects.

Role of the left posterior middle temporal gyrus in shape recognition and its reconstruction during drawing: A study combining transcranial magnetic stimulation and functional near infrared spectroscopy.

There are numerous reports of enhanced or emerged visual arts abilities in patients with semantic impairment. These reports led to the theory that a loss of function on the language side of the brain can result in changes of ability to draw and/or to paint. Further, the left posterior middle temporal gyrus (l-pMTG) has been revealed to contribute to the higher control semantic mechanisms with objects recognition and integration of visual information, within a widely distributed network of the left hemisphere. Nevertheless, the theory has not been fully studied in neural bases. The aim of this study is to examine role of the l-pMTG on shape recognition and its reconstruction within drawing behavior, by using a combining method of the repetitive transcranial magnetic stimulation (rTMS) and functional near-infrared spectroscopy (fNIRS). Eighteen healthy participants received a low frequency inhibitory rTMS to their l-pMTG during the drawing task of the Benton Visual Retention Test (BVRT). There was a significant decrease of the mean accuracy of reproductions in the Complex designs of the BVRT, compared to the Simple and Medium designs. The fNIRS data showed strong negative correlations with the results of the BVRT. Though our hypothesis had a contradiction that rTMS would have inhibited the brain activity in the stimulated site, the results suggest that shape recognition and its reconstruction such as the BVRT require neural activations of the l-TL as well as that of the l-pMTG.

A novel nondestructive detection approach for seed cotton lint percentage using deep learning

BackgroundThe lint percentage of seed cotton is one of the most important parameters for evaluating seed cotton quality and affects its price. The traditional measuring method of lint percentage is labor-intensive and time-consuming; thus, an efficient and accurate measurement method is needed. In recent years, classification-based deep learning and computer vision have shown promise in solving various classification tasks.ResultsIn this study, we propose a new approach for detecting the lint percentage using MobileNetV2 and transfer learning. The model is deployed on a lint percentage detection instrument, which can rapidly and accurately determine the lint percentage of seed cotton. We evaluated the performance of the proposed approach using a dataset comprising 66 924 seed cotton images from different regions of China. The results of the experiments showed that the model with transfer learning achieved an average classification accuracy of 98.43%, with an average precision of 94.97%, an average recall of 95.26%, and an average F1-score of 95.20%. Furthermore, the proposed classification model achieved an average accuracy of 97.22% in calculating the lint percentage, showing no significant difference from the performance of experts (independent-sample t-test, t = 0.019, P = 0.860).ConclusionThis study demonstrated the effectiveness of the MobileNetV2 model and transfer learning in calculating the lint percentage of seed cotton. The proposed approach is a promising alternative to traditional methods, providing a rapid and accurate solution for the industry.

Enhancing additive manufacturing with computer vision: a comprehensive review

Enhancing additive manufacturing with computer vision: a comprehensive review

A novel deep learning based CBIR model using Convolutional Siamese Neural Networks

Content-Based Image Retrieval (CBIR) is a technique that involves retrieving similar images from a large database by analysing the content features of the query image. The heavy usage of digital platforms and devices has in a way promoted CBIR and its allied technologies in computer vision and artificial intelligence. The process entails comparing the representative features of the query image with those of the images in the dataset to rank them for retrieval. Past research was centered around handcrafted feature descriptors based on traditional visual features. But with the advent of deep learning the traditional manual method of feature engineering gave way to automatic feature extraction. In this study, a cascaded network is utilised for CBIR. In the first stage, the model employs multi-modal features from variational autoencoders and super-pixelated image characteristics to narrow down the search space. In the subsequent stage, an end-to-end deep learning network known as a Convolutional Siamese Neural Network (CSNN) is used. The concept of pseudo-labeling is incorporated to categorise images according to their affinity and similarity with the query image. Using this pseudo-supervised learning approach, this network evaluates the similarity between a query image and available image samples. The Siamese network assigns a similarity score to each target image, and those that surpass a predefined threshold are ranked and retrieved. The suggested CBIR system undergoes testing on a widely recognized public dataset: the Oxford dataset and its performance is measured against cutting-edge image retrieval methods. The findings reveal substantial enhancements in retrieval performance in terms of several standard benchmarks such as average precision, average error rate, average false positive rate etc., providing strong support for utilising images from interconnected devices.

Smartphone Camera for Angiographic Computer Vision in Vascular Medicine

Aim: This study aimed to develop a TensorFlow Lite algorithm for angiography classification and to deploy it on a basic mobile smartphone device, thereby verifying the proof of concept for creating a comprehensive end-to-end mobile computer vision application for vascular medicine. Materials and Methods: After ethical approval by the local ethics committee, we collected institutional and open source peripheral angiograms of lower limbs. The angiograms were labeled by a researcher with more than 10 years of experience in vascular surgery. The labeling included dividing the angiograms according to their anatomical pattern into the Global Limb Anatomic Staging System (GLASS). The model was developed using the open-source TensorFlow framework for general image classification and deployed as an Android application. Results: The model utilized 700 angiograms, distributed as follows within the femoropoliteal GLASS disease (fp) categories: fp0 – 187 images, fp1 – 136 images, fp2 – 128 images, fp3 – 97 images, fp4 – 152 images. The reference dataset included 372 non-angiographic images (not_angio). Consequently, the entire model included 1,072 images. After training and deployment, the model demonstrated the following performance: a mean accuracy of 0.72. The best self-reported accuracy per class was for fp0 0.72, fp4 0.83 and not_angio 1.0 classes. Conclusion: We discovered that a smartphone camera could be utilized for angiographic computer vision through end-to-end applications accessible to every healthcare professional. However, the predictive abilities of the model are limited and require improvement. The development of a robust angiographic computer vision smartphone application should incorporate an upload function, undergo validation through head-to-head human-machine comparisons, potentially include segmentation, and feature a prospective design with explicit consent for using collected data in the development of AI models.