Object Identification System Research Articles

AUTOMATIC PANEL QUALITY MONITORING SYSTEM USING VISION-BASED ABNORMAL OBJECT DETECTION

The rate of “Not Good” products due to the presence of abnormal objects during the quality control inspection of mobile phone’s panels is as high as 38 percent without supervision and 19 percent with naked eye’s inspection. To reduce the loss of small abnormal objects and improve the recognition rate of object detection of automatic panel quality monitoring, the paper proposes abnormal object recognition based on a template matching and subtract background technique. Firstly, the object recognition network based on convolutional neural networks (CNNs) is introduced initially. Then, the Canny Edge Detection method improves the image input quality. Next, template matching is utilized to discover things that depart from a standard pattern or a predetermined model. Models for background reduction will be implemented instantly into smartphone panel quality checking systems. Continually, a vision-based quality control (QC) strategy will be developed. Lastly, experimental results from a practical vision-based automatic anomalous object identification system demonstrate the viability and enhancement of the suggested strategy.

A new CNN-BASED object detection system for autonomous mobile robots based on real-world vehicle datasets

Recently, autonomous mobile robots (AMRs) have begun to be used in the delivery of goods, but one of the biggest challenges faced in this field is the navigation system that guides a robot to its destination. The navigation system must be able to identify objects in the robot's path and take evasive actions to avoid them. Developing an object detection system for an AMR requires a deep learning model that is able to achieve a high level of accuracy, with fast inference times, and a model with a compact size that can be run on embedded control systems. Consequently, object recognition requires a convolutional neural network (CNN)-based model that can yield high object classification accuracy and process data quickly. This paper introduces a new CNN-based object detection system for an AMR that employs real-world vehicle datasets. First, we create original real-world datasets of images from Banda Aceh city. We then develop a new CNN-based object identification system that is capable of identifying cars, motorcycles, people, and rickshaws under morning, afternoon, and evening lighting conditions. An SSD Mobilenetv2 FPN Lite 320 × 320 architecture is employed for retraining using these real-world datasets. Quantitative and qualitative performance indicators are then applied to evaluate the CNN model. Training the pre-trained SSD Mobilenetv2 FPN Lite 320 × 320 model improves its classification and detection accuracy, as indicated by its performance results. We conclude that the proposed CNN-based object detection system has the potential for use in an AMR.

Efficient Object Detection with YOLO: A Comprehensive Guide

Object detection presents itself as a pivotal and complex challenge within the domain of computer vision. Over the past ten years, as deep learning techniques have advanced quickly, researchers have committed significant resources to utilising deep models as the basis to improve the performance of object identification systems and related tasks like segmentation, localization. Two-stage and single-stage detectors are the two basic categories into which object detectors can be roughly divided. Typically, two-stage detectors use complicated structures in conjunction with a selective region proposal technique to accomplish their goals. Conversely, single-stage detectors aim to detect objects across all spatial regions in one shot, employing relatively simpler architectures. Any object detector's inference time and detection accuracy are the main factors to consider while evaluating it. Single-stage detectors offer quicker inference times, but two-stage detectors frequently show better detection accuracy. But since the introduction of YOLO (You Only Look Once) and its architectural offspring, detection accuracy has significantly improved—sometimes even outperforming that of two-stage detectors. The adoption of YOLO in various applications is primarily driven by its faster inference times rather than its detection accuracy alone.

An Efficient Object Detection System for the Blind People

As object recognition technology has developed recently, various technologies have been applied to autonomous vehicles, robots, and industrial facilities. However, the benefits of these technologies are not reaching the visually impaired, who need it the most. In this research, researchers proposed a deep learning based on object identification system for the visually impaired. Voice recognition technology is used to know what objects a blind person wants, and then to find the objects via object recognition. Furthermore, a voice guidance technique is used to inform sight impaired persons as to the location of objects. The object recognition deep learning model utilizes the single shot multi-box detector (SSD) neural network architecture, and voice recognition is designed through speech-to-text (STT) technology. In addition, a voice announcement is synthesized using text-to-speech (TTS) to make it easier for the blind to get information about objects. The system is built using python OpenCV tool. As a result, we implement an efficient object-detection system that helps the blind find objects in a specific space without help from others, and the system is analyse through experiments to verify performance. Keywords: Object detection system, blind people, Open CV, Voice Recognition

A Multimodal Fusion System for Object Identification in Point Clouds with Density and Coverage Differences

Data fusion, which involves integrating information from multiple sources to achieve a specific objective, is an essential area of contemporary scientific research. This article presents a multimodal fusion system for object identification in point clouds in a controlled environment. Several stages were implemented, including downsampling and denoising techniques, to prepare the data before fusion. Two denoising approaches were tested and compared: one based on neighborhood technique and the other using a median filter for each “x”, “y”, and “z” coordinate of each point. The downsampling techniques included Random, Grid Average, and Nonuniform Grid Sample. To achieve precise alignment of sensor data in a common coordinate system, registration techniques such as Iterative Closest Point (ICP), Coherent Point Drift (CPD), and Normal Distribution Transform (NDT) were employed. Despite facing limitations, variations in density, and differences in coverage among the point clouds generated by the sensors, the system successfully achieved an integrated and coherent representation of objects in the controlled environment. This accomplishment establishes a robust foundation for future research in the field of point cloud data fusion.

BUILDING A MOVING OBJECT IDENTIFICATION SYSTEM BASED ON MACHINE LEARNING TECHNOLOGIES

The study is devoted to the construction of a system for identifying moving objects in a video stream based on machine learning technologies. Tracking and recognizing moving objects is an urgent task of our time. It is important to recognize objects in motion and identify them based on artificial intelligence. The system is divided into three main modules: face recognition, people tracking, and saving of recognition results. The use of modern technologies and YOLOv7 machine learning algorithms for tracking people and the Face Recognition library for face recognition is described. A contextual Data flow diagram is created, which shows the sequence of steps required to convert the input video stream into normalized face images that are ready for further recognition. The hierarchy of processes of the moving object identification system is built. The video processing process decomposition diagram shows the logical sequence of stages and data flows required to prepare face images. Behavior classification associates detected motion patterns with specific types of behavior. The system uses facial identification data and information about their previous behavior to classify movement patterns. The process decomposition allowed us to consider in detail each of the key aspects of the system and reveal the sequence of steps and data flows required for their implementation. Building a process hierarchy diagram made it possible to qualitatively display the relationships between all processes and subprocesses of the system, demonstrating the logical sequence of their execution. The ER diagram defined the structure of the database used to store information about individuals. The system analysis laid the foundation for the further design and development of the information system for tracking and recognizing people. It allowed us to determine the main functional requirements, the structure of the system and the relationships between its components. particular importance is the ability to use the software to prevent terrorist and sabotage threats. Thanks to such information systems, it is possible to improve the economic situation of both individual facilities and the country as a whole.

AI-based outdoor moving object detection for smart city surveillance

<abstract> <p>One essential component of the futuristic way of living in "smart cities" is the installation of surveillance cameras. There are a wide variety of applications for surveillance cameras, including but not limited to: investigating and preventing crimes, identifying sick individuals (coronavirus), locating missing persons, and many more. In this research, we provided a system for smart city outdoor item recognition using visual data collected by security cameras. The object identification model used by the proposed outdoor system was an enhanced version of RetinaNet. A state of the art object identification model, RetinaNet boasts lightning-fast processing and pinpoint accuracy. Its primary purpose was to rectify the focal loss-based training dataset's inherent class imbalance. To make the RetinaNet better at identifying tiny objects, we increased its receptive field with custom-made convolution blocks. In addition, we adjusted the number of anchors by decreasing their scale and increasing their ratio. Using a mix of open-source datasets including BDD100K, MS COCO, and Pascal Vocab, the suggested outdoor object identification system was trained and tested. While maintaining real-time operation, the suggested system's performance has been markedly enhanced in terms of accuracy.</p> </abstract>

Dependable Cyber-Physical Systems: concepts, challenges, and case studies

As the integration of cyber and physical components becomes increasingly prevalent in modern systems, ensuring the dependability of Cyber-Physical Systems (CPS) has emerged as a critical research area. The present paper aims to present a comprehensive analysis of the recent advances in the field of dependable CPS. Firstly, we present an analysis on the evolution of the published papers in the past years and introduce the concepts of dependability and CPSs in the context of Industry 4.0. Secondly, some strategies and challenges that have been encountered in the recent literature for enhancing the dependability of CPS are also presented. Finally, two case studies that highlight the interdependencies between the concepts of dependability and CPS are reviewed. The first one presents the dependability of an object identification system that relies on radio frequency identifier, which can be used in medical services. The second case study shows various techniques for predicting the remaining useful life of a turbofan jet engine using Machine Learning and Deep Learning techniques.

Research on algorithms based on object detection

Object detection technology is a hot research direction in computer vision field technology, which is broadly used in face recognition, vehicle navigation, aviation and other important fields, with broad development prospects. In recent years, object detection algorithms based on deep learning have also appeared as computer science technology has advanced quickly. Comparing modern object detecting algorithms to conventional ones, this algorithm has gradually highlighted the advantages of high precision and good real-time performance. This article reviews traditional object detection algorithms, and focuses on the HOG algorithm of traditional object detection, reviews deep learning-based two-stage and one-stage object identification systems and weighs the benefits and drawbacks of each. A summary and outlook for the future object detection algorithm development is also provided.

A Comparative Study of Object Detection using YOLO and SSD Algorithms

One of the most significant and difficult areas of computer vision is object detection, which is used in many aspects of daily life, including security monitoring, autonomous vehicle operation, and so on, to find instances of semantic objects belonging to a particular class. In this survey, we first look at the current approaches of typical detection models and discuss the benchmark datasets in order to comprehensively and deeply comprehend the major development status of the object detection pipeline. This project explains how convolutional neural network-based deep learning techniques are used for object detection. In this study, deep learning methods for cutting-edge object identification systems are evaluated. Any kind of visual medium is represented by a computer as a collection of numerical numbers. In order to inspect the contents of images, they need image processing techniques. In order to determine which is the quickest and most effective object detection algorithm, this research analyzes two popular algorithms: Single Shot Detection (SSD) and You Only Look Once (YOLO). The COCO (Common Object in Context) dataset is used in this comparative research to assess the effectiveness of the two algorithms and to analyze their strengths and weaknesses in terms of factors like accuracy, precision, and speed. Based on the findings analysis, it can be said that YOLO-v3 outperforms SSD in the same testing scenario. Keywords- Object Detection, Deep learning, CNN, COCO, YOLO, SSD

ANALYSIS OF INFORMATION TECHNOLOGIES FOR REMOTE IDENTIFICATION OF DYNAMIC OBJECTS

The problem of identification of dynamic objects using remote identification information technologies is considered. It is noted that the identification of moving objects is important in various fields, including autonomous vehicles, medical diagnostics and robotics. The purpose of the article is to analyze various information technologies for detecting objects that can be used in future research on remote identification. Analysis of methods for determining speed as a dynamic parameter, analysis of two-step and one-step methods of remote identification of objects, analysis of early identification methods, as well as analysis of methods for improving remote identification of objects was carried out. Several means of determining the motion of objects are considered, in particular, the proportional-integral-differential controller, the leveling block method, phase correlation, pixel recursion algorithms, and the optical flow methods of Lucas – Kanady, Horn – Shunk, Farnbeck, dense optical flow. These tools can be used to effectively determine the movement of objects and identify their speed regardless of the size and position of the objects. Two-step and one-step object detection methods are considered: region method with convolutional neural networks, its improvements, spatial pyramid pooling networks, "You only look once" method, one-step multi-frame method, retinal networks, corner network, central network and detection transformer, which use different approaches to improve the performance and accuracy of object detection. The necessity of using methods of convolutional neural networks and spatial pyramid pooling networks for effective identification of objects regardless of their size and position is emphasized. New approaches are proposed that allow creating fixed-length representations for image processing and regions of interest, as well as Viola – Jones methods, oriented gradient histograms, and deformed part models. Research in the field of object detection contributes to the development of information technologies and the improvement of the efficiency of dynamic object identification systems. Through the review and analysis of various methods, recommendations for researchers and practitioners working in the field of remote identification of dynamic objects are provided.

Research advanced in object detection based on deep learning

In the field of computer vision, identifying specific objects in images and predicting their location and class have long been hot research topics.Algorithms for early object detection rely on traditional handcrafted features, and the speed and precision of their detection cannot meet the needs of real-world applications. Rapid development in convolutional neural networks has accelerated the development of object identification systems based on deep learning.Existing deep learning-based object identification algorithms mostly use two-stage detection and single-stage detection, according to various detection frameworks. In this paper, around the above two types of frameworks, the latest research progress in the field of object detection is systematically introduced. Specifically, we first introduce representative object detection algorithms, including their design ideas, basic processes, and advantages and disadvantages. Second, using widely-used datasets, we objectively compare the effectiveness of several detection techniques. Finally, we summarize the unsolved problems in the detection of objects and talk about the route this topic will take going forward.

Ventral and Dorsal Stream EEG Channels: Key Features for EEG-Based Object Recognition and Identification.

Object recognition and object identification are multifaceted cognitive operations that require various brain regions to synthesize and process information. Prior research has evidenced the activity of both visual and temporal cortices during these tasks. Notwithstanding their similarities, object recognition and identification are recognized as separate brain functions. Drawing from the two-stream hypothesis, our investigation aims to understand whether the channels within the ventral and dorsal streams contain pertinent information for effective model learning regarding object recognition and identification tasks. By utilizing the data we collected during the object recognition and identification experiment, we scrutinized EEGNet models, trained using channels that replicate the two-stream hypothesis pathways, against a model trained using all available channels. The outcomes reveal that the model trained solely using the temporal region delivered a high accuracy level in classifying four distinct object categories. Specifically, the object recognition and object identification models achieved an accuracy of 89% and 85%, respectively. By incorporating the channels that mimic the ventral stream, the model's accuracy was further improved, with the object recognition model and object identification model achieving an accuracy of 95% and 94%, respectively. Furthermore, the Grad-CAM result of the trained models revealed a significant contribution from the ventral and dorsal stream channels toward the training of the EEGNet model. The aim of our study is to pinpoint the optimal channel configuration that provides a swift and accurate brain-computer interface system for object recognition and identification.

Lightweight and Parameter-Optimized Real-Time Food Calorie Estimation from Images Using CNN-Based Approach

Automated object identification has seen significant progress during the last decade with close to human-level accuracy, aided by deep learning methods. With the rapid rise of obesity and other lifestyle-related diseases worldwide, the availability of fast, automated, and reliable image-based food calorie estimation is becoming a necessity. With the help of a deep learning-based automated object identification system, it is possible to introduce accurate and intelligent solutions in the form of a mobile app. However, for these kind of applications, processing speed is an important concern as the images should be processed in real time. Although plenty of studies have been conducted that focus on food image detection-based calorie estimation, there is still a lack of an image-driven, lightweight, fast, and reliable food calorie estimation system. In this paper, we propose a method based on the parameter-optimized Convolution Neural Networks (CNN) for detecting food images of regular meals using a handheld camera. Once identification process of the food items are complete, the corresponding calories and nutritional facts can be calculated using prior knowledge about the food class. Through our findings, we demonstrate that our proposed approach ensures high accuracy and can significantly simplify the existing manual calorie estimation procedures by converting them into a real-time automated process.

Deep Learning based Object Detection Techniques in Videos

Deep learning technology is often used for object detection. It has received attention recently because to the intimate connections between object detection, video analysis, and picture understanding. The goal of object detection has been pursued using a variety of models, and this is immensely beneficial to humanity. The most recent technical developments have helped the computational experiments, which would not have been conceivable if they had been tried using the conventional techniques. The powerful approaches employed in deep learning can show noticeably higher efficiency when compared to conventional designs and architectures. Numerous strategies and techniques have been used in deep learning to boost accuracy, and their drawbacks have also been somewhat addressed in order to lessen them. This study's main objective is to give an overview of several object detection procedures and approaches based on deep learning. Additionally, it lists the benefits and drawbacks of various object identification systems based on their potential applications and limitations.

Multiwavelength Optoelectronic Synapse with 2D Materials for Mixed-Color Pattern Recognition.

Neuromorphic visual systems emulating biological retina functionalities have enormous potential for in-sensor computing, with prospects of making artificial intelligence ubiquitous. Conventionally, visual information is captured by an image sensor, stored by memory units, and eventually processed by the machine learning algorithm. Here, we present an optoelectronic synapse device with multifunctional integration of all the processes required for real time object identification. Ultraviolet-visible wavelength-sensitive MoS2 FET channel with infrared sensitive PtTe2/Si gate electrode enables the device to sense, store, and process optical data for a wide range of the electromagnetic spectrum, while maintaining a low dark current. The device exhibits optical stimulation-controlled short-term and long-term potentiation, electrically driven long-term depression, synaptic weight update for multiple wavelengths of light ranging from 300 nm in ultraviolet to 2 μm in infrared. An artificial neural network developed using the extracted weight update parameters of the device can be trained to identify both single wavelength and mixed wavelength patterns. This work demonstrates a device that could potentially be used for realizing a multiwavelength neuromorphic visual system for pattern recognition and object identification.

Adaptive Authentication Protocol Based on Zero-Knowledge Proof

Authentication protocols are expanding their application scope in wireless information systems, among which are low-orbit satellite communication systems (LOSCS) for the OneWeb space Internet, automatic object identification systems using RFID, the Internet of Things, intelligent transportation systems (ITS), Vehicular Ad Hoc Network (VANET). This is due to the fact that authentication protocols effectively resist a number of attacks on wireless data transmission channels in these systems. The main disadvantage of most authentication protocols is the use of symmetric and asymmetric encryption systems to ensure high cryptographic strength. As a result, there is a problem in delivering keys to the sides of the prover and the verifier. At the same time, compromising of keys will lead to a decrease in the level of protection of the transmitted data. Zero-knowledge authentication protocols (ZKAP) are able to eliminate this disadvantage. However, most of these protocols use multiple rounds to authenticate the prover. Therefore, ZKAP, which has minimal time costs, is developed in the article. A scheme for adapting protocol parameters has been developed in this protocol to increase its efficiency. Reductions in the level of confidentiality allow us to reduce the time spent on the execution of the authentication protocol. This increases the volume of information traffic. At the same time, an increase in the confidentiality of the protocol entails an increase in the time needed for authentication of the prover, which reduces the volume of information traffic. The FPGA Artix-7 xc7a12ticsg325-1L was used to estimate the time spent implementing the adaptive ZKAP protocol. Testing was performed for 32- and 64-bit adaptive authentication protocols.

Velostat Sensor Array for Object Recognition

This paper presents a cost-effective pressure sensing system for object detection and identification. The pressure sensing system consists of a <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$27\times 27$ </tex-math></inline-formula> piezoresistive sensor array made of carbon composite Velostat, a signal processing subsystem for signal scanning, amplification, registration, and enhancement. A convolutional neural network is used to classify various objects through the pressure signals produced and processed by the sensing array. Based on systematic characterizations and calibrations of sensing materials and system sensitivity, three experiment setups are established to recognize 10 objects to be detected. In series of experiments, a pressure image data set consisting of 32264 frames of images is first assembled to represent the 10 objects. Contrast enhancement algorithm was used to process the pressure image data set and combined with a convolutional neural network ResNet-PI to classify the 10 objects. For pressure images collected with the preestablished three experiment setups, an overall accuracy of 0.9854 is achieved. Compared with other systems based on Velostat sensor array, the system demonstrated in this study features improvements in structural robustness, detection repeatability and system reliability, suggesting its potential applications in emerging areas including human-computer interaction and smart health monitoring.

Recognition of Image-Based Plant Leaf Diseases Using Deep Learning Classification Models

Plant diseases are spread by a variety of pests, weeds, and pathogens and may have a devastating effect on agriculture, if not handled in a timely manner. Farmers face umpteen challenges from a proper water supply, untimely rain, storage facilities, and several plant diseases. Crops disease is the primary threat and it causes enormous loss to farmers in terms of production and finance. Identifying the disease from several hectares of agricultural land is a very difficult practice even with the presence of modern technology. Accurate and rapid illness prediction for early illness treatment to crops minimizes economical loss to the individual and further proves to be productive for healthy crops. Many studies use modern deep learning approaches to improve the accuracy and performance of object detection and identification systems. The suggested method notifies farmers of different agricultural illnesses, prompting them to take further essential precautions before the disease spreads to the whole agricultural field. The primary objective of this study is to detect the illnesses as soon as they begin to spread on the leaves of the plants. Super-Resolution Convolutional Neural Network (SRCNN) and Bicubic models are employed in the system to identify healthy and diseased leaves with an accuracy of 99.175 % and 99.156 % respectively.

A Deep Learning Based Object Identification System for Forest Fire Detection

Forest fires are still a large concern in several countries due to the social, environmental and economic damages caused. This paper aims to show the design and validation of a proposed system for the classification of smoke columns with object detection and a deep learning-based approach. This approach is able to detect smoke columns visible below or above the horizon. During the dataset labelling, the smoke object was divided into three different classes, depending on its distance to the horizon, a cloud object was also added, along with images without annotations. A comparison between the use of RetinaNet and Faster R-CNN was also performed. Using an independent test set, an F1-score around 80%, a G-mean around 80% and a detection rate around 90% were achieved by the two best models: both were trained with the dataset labelled with three different smoke classes and with augmentation; Faster R-CNNN was the model architecture, re-trained during the same iterations but following different learning rate schedules. Finally, these models were tested in 24 smoke sequences of the public HPWREN dataset, with 6.3 min as the average time elapsed from the start of the fire compared to the first detection of a smoke column.