Object Detection Applications Research Articles

High-strength synergic-calibration attention system in YOLO for underwater object detection application

High-strength synergic-calibration attention system in YOLO for underwater object detection application

Realization of Secure Robotic Brain Via Programmable Metasurface with Robust High‐Order BIC

AbstractTerahertz waves can be widely used for short‐range communication in complex indoor environments and non‐destructive object detection applications. Metasurfaces are widely used in terahertz sensing and communication devices because they can modulate terahertz waves in multiple dimensions. Metamaterial robot brain can utilize metasurfaces' powerful direct modulation ability to achieve sensing and communication functions. The metasurface devices realized based on Dynamic Heterogeneous Redundancy (DHR) architecture can improve the confidentiality and security of terahertz wave wireless communication. While the intrinsic ohmic loss and quality factor of usual metallic metamaterials are usually low, the concept of bound states in the continuum (BIC) has been proposed for stronger terahertz‐matter interactions. Among them, high‐order BICs are of interest because of their strong robustness to structural defects. Therefore, an aluminium‐graphene hybrid metasurface with high‐order BIC is proposed. We have the principle of excitation of high‐order BICs is investigated and creatively proposed with high robustness realized using the magnetical EIT effect. The robustness of the high‐order BIC is also utilized to design security hardware based on DHR architecture. The designed secure hardware can satisfy the demand for an intelligent robotic brain to the internal terahertz wave confidential wireless communication.

Investigation and Application of Various Algorithms used in Object Detection and Classification in Image Data

In this article, we compare and analyze the YOLO (You Only Look Once) method, widely employed for object detection in digital image processing, with the Haar feature-based cascade classifier method implemented using the OpenCV library. YOLO, a deep learning–based approach, excels in real-time object detection and recognition applications. In contrast, the Haar method utilizes a traditional approach to rapidly identify features. However, significant performance differences exist between the two methods. Experimental results and performance analyses demonstrate that YOLO provides high accuracy rates and real-time processing speeds in object detection tasks. The code implementations presented in this study will be valuable to researchers new to digital image processing. Additionally, YOLO has shown high performance on large and complex datasets by leveraging GPU capabilities. Experiments with various YOLO versions (e.g., YOLOv4, YOLOv5, YOLOv7) have established it as one of the most suitable options for real-time applications, particularly due to its low latency and high accuracy.

Cross-pollination of knowledge for object detection in domain adaptation for industrial automation

AbstractArtificial Intelligence is revolutionizing industries by enhancing efficiency through real-time Object Detection (OD) applications. Utilizing advanced computer vision techniques, OD systems automate processes, analyze complex visual data, and facilitate data-driven decisions, thus increasing productivity. Domain Adaptation for OD has recently gained prominence for its ability to recognize target objects without annotations. Innovative approaches that merge traditional cross-disciplinary domain modeling with cutting-edge deep learning have become essential in addressing complex AI challenges in real-time scenarios. Unlike traditional methods, this study proposes a novel, effective Cross-Pollination of Knowledge (CPK) strategy for domain adaptation inspired by botanical processes. The CPK approach involves merging target samples with source samples at the input stage. By incorporating a random and unique selection of a few target samples, the merging process enhances object detection results efficiently in domain adaptation, supporting detectors in aligning and generalizing features with the source domain. Additionally, this work presents the new Planeat digit recognition dataset, which includes 231 images. To ensure robust comparison, we employ a self-supervised Domain Adaptation (UDA) method that simultaneously trains target and source domains using unsupervised techniques. UDA method leverages target data to identify high-confidence regions, which are then cropped and augmented, adapting UDA for effective OD. The proposed CPK approach significantly outperforms existing UDA techniques, improving mean Average Precision (mAP) by 10.9% through rigorous testing on five diverse datasets across different conditions- cross-weather, cross-camera, and synthetic-to-real. Our code is publicly available https://github.com/anwaar0/CPK-Object-Detection

Green fruit detection methods: Innovative application of camouflage object detection and multilevel feature mining

Accurate detection of object fruits is essential for optimizing picking efficiency and predicting fruit yields. However, detecting early-stage ripening or green fruits, especially in complex fields, poses significant challenge due to their similarity to green leaves. This study introduces the TEAVit model, a novel camouflage object detection network specifically tailored for identifying green tomatoes in intricate agricultural environments. TEAVit incorporates a texture-edge-awareness module (TEAM) to enhance the extraction ability of texture feature by combining high-level and low-level features, an edge-guided feature module (EFM) to address background complexities and occlusions, and a context-aggregation module (CAM) to leverage contextual semantics. Experimental validation results demonstrate that the S-measure, E-measure, and F-measure performance metrics all exceed 90% on the tomato dataset, accompanied by a mean absolute error of 0.0245. These findings underpinned the effectiveness of the proposed green fruit camouflage object detection algorithm in offering new insights for agricultural target localization.

Enhancing robustness of person detection: A universal defense filter against adversarial patch attacks

Person detection is one of the most popular object detection applications, and has been widely used in safety-critical systems such as autonomous driving. However, recent studies have revealed that person detectors are vulnerable to physically adversarial patch attacks and may suffer detection failure. Data-side defense is an effective approach to address this issue, owing to its low computational cost and ease of deployment. However, existing data-side defenses have limited effectiveness in resisting adaptive patch attacks. To overcome this challenge, we propose a new data-side defense, called Universal Defense Filter (UDFilter). UDFilter covers the input images with an equal-size defense filter to weaken the negative impact of adversarial patches. The defense filter is generated using a self-adaptive learning algorithm that facilitates iterative competition between adversarial patch and defense filter, thus bolstering UDFilter’s ability to defense adaptive attacks. Furthermore, to maintain the clean performance, we propose a plug-and-play Joint Detection Strategy (JDS) during the model testing phase. Extensive experiments have shown that UDFilter can significantly enhance robustness of person detection against adversarial patch attacks. Moreover, UDFilter does not result in a discernible reduction in the model’s clean performance.

Recent Advances for Aerial Object Detection: A Survey

Aerial object detection, as object detection in aerial images captured from an overhead perspective, has been widely applied in urban management, industrial inspection, and other aspects. However, the performance of existing aerial object detection algorithms is hindered by variations in object scales and orientations attributed to the aerial perspective. This survey presents a comprehensive review of recent advances in aerial object detection. We start with some basic concepts of aerial object detection and then summarize the five imbalance problems of aerial object detection, including scale imbalance, spatial imbalance, objective imbalance, semantic imbalance, and class imbalance. Moreover, we classify and analyze relevant methods and especially introduce the applications of aerial object detection in practical scenarios. Finally, the performance evaluation is presented on two popular aerial object detection datasets VisDrone-DET and DOTA, and we discuss several future directions that could facilitate the development of aerial object detection.

Online_XKD: An online knowledge distillation model for underwater object detection

Underwater object detection in the field of computer vision faces unique challenges such as color distortion, reduced visibility, and blurred edges caused by aquatic conditions, which decrease the efficiency of traditional methods. Furthermore, the limited memory and computational power of underwater equipment necessitate the development of lightweight and efficient real-time monitoring algorithms. To address these issues, this paper designs an online knowledge distillation(KD) framework based on mutual knowledge transfer, named Online-XKD, which aims to enhance the efficiency and generalizability of knowledge distillation through mutual knowledge exchange while maintaining model lightweightness. The core idea of Online-XKD is an internally designed mutual knowledge transfer structure for online distillation within the detection head, which enhances learning outcomes by optimizing distillation efficiency and broadening generalizability. Additionally, we introduce a mask generation feature knowledge distillation architecture based on the backbone network, which significantly enhances the feature extraction capability of the student model by integrating mask masking and feature generation techniques. Moreover, by incorporating PSA and AUGFPN attention modules, the feature extraction capability in complex underwater environments is further enhanced without adding extra parameters, thereby improving detection accuracy. Experimental results on the URPC2020 dataset demonstrate that Online-XKD enhances the detection accuracy of the student model by 3.6 mAP, exceeding initial KD models by 3.3 mAP, and outperforming most existing KD methods in underwater detection tasks. These results confirm the effectiveness and superiority of Online-XKD in applications of underwater object detection, proving its practical significance and effectiveness in enhancing the feature extraction and generalization abilities of lightweight models.

A High-Performance and Ultra-Low-Power Accelerator Design for Advanced Deep Learning Algorithms on an FPGA

This article addresses the growing need in resource-constrained edge computing scenarios for energy-efficient convolutional neural network (CNN) accelerators on mobile Field-Programmable Gate Array (FPGA) systems. In particular, we concentrate on register transfer level (RTL) design flow optimization to improve programming speed and power efficiency. We present a re-configurable accelerator design optimized for CNN-based object-detection applications, especially suitable for mobile FPGA platforms like the Xilinx PYNQ-Z2. By not only optimizing the MAC module using Enhanced clock gating (ECG), the accelerator can also use low-power techniques such as Local explicit clock gating (LECG) and Local explicit clock enable (LECE) in memory modules to efficiently minimize data access and memory utilization. The evaluation using ResNet-20 trained on the CIFAR-10 dataset demonstrated significant improvements in power efficiency consumption (up to 22%) and performance. The findings highlight the importance of using different optimization techniques across multiple hardware modules to achieve better results in real-world applications.

Development and challenges of object detection: A survey

Object detection is a basic vision task that accompanies people’s daily lives all the time. The development of object detection technology has experienced an evolution from traditional-based algorithms to deep learning-based algorithms, which has made a qualitative leap in both detection accuracy and detection speed. With the advancement of deep learning, object detection techniques are increasingly becoming a part of everyday life, with the YOLO series of algorithms being extensively applied in various industries. In this paper, we initially present the frequently utilized datasets and evaluation criteria for object detection. Subsequently, we delve into the evolution of traditional object detection algorithms, highlighting two-stage and one-stage approaches through illustrative examples of classical methods. We also conduct a comprehensive summary and analysis of the detection results obtained by these methods. In addition, we introduce object detection applications in daily life, as well as the importance and some difficulties of these applications. Finally, we analyze and summarize the difficulties and challenges facing the task of object detection, and we look forward to the future development direction of object detection.

LRMSNet: A New Lightweight Detection Algorithm for Multi-Scale SAR Objects

In recent years, deep learning has found widespread application in SAR image object detection. However, when detecting multi-scale targets against complex backgrounds, these models often struggle to strike a balance between accuracy and speed. Furthermore, there is a continuous need to enhance the performance of current models. Hence, this paper proposes LRMSNet, a new multi-scale target detection model designed specifically for SAR images in complex backgrounds. Firstly, the paper introduces an attention module designed to enhance contextual information aggregation and capture global features, which is integrated into a backbone network with an expanded receptive field for improving SAR image feature extraction. Secondly, this paper develops an information aggregation module to effectively fuse different feature layers of the backbone network. Lastly, to better integrate feature information at various levels, this paper designs a multi-scale aggregation network. We validate the effectiveness of our method on three different SAR object detection datasets (MSAR-1.0, SSDD, and HRSID). Experimental results demonstrate that LRMSNet achieves outstanding performance with a mean average accuracy (mAP) of 95.2%, 98.9%, and 93.3% on the MSAR-1.0, SSDD, and HRSID datasets, respectively, with only 3.46 M parameters and 12.6 G floating-point operation cost (FLOPs). When compared with existing SAR object detection models on the MSAR-1.0 dataset, LRMSNet achieves state-of-the-art (SOTA) performance, showcasing its superiority in addressing SAR detection challenges in large-scale complex environments and across various object scales.

Dynamic YOLO for small underwater object detection

The practical application of object detection inevitably encounters challenges posed by small objects. In underwater object detection, a crucial method for marine exploration, the presence of small objects in underwater environments significantly hampers the performance of detection. In this paper, a dynamic YOLO detector is proposed as a solution to alleviate this problem. Specifically, a light-weight backbone network is first constructed based on deformable convolution v3, with some specialized designs for small object detection. Secondly, a unified feature fusion framework based on channel-wise, scale-wise, and spatial-aware attention is proposed to fuse feature maps from different scales. This is particularly critical for detecting small objects since it allows us to fully exploit the enhanced capabilities offered by our proposed backbone network. Finally, a simple but effective detection head is designed to handle the conflict between classification and localization by disentangling and aligning the two tasks. Extensive experiments are conducted on benchmark datasets to demonstrate the effectiveness of the proposed model. Without bells and whistles, dynamic YOLO outperforms the recent state-of-the-art methods by a large margin of +0.8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$+\\,0.8$$\\end{document} AP and +1.8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$+\\,1.8$$\\end{document} APS\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ ext {AP}_{S}$$\\end{document} on the DUO dataset. Experimental results on Pascal VOC and MS COCO datasets also demonstrate the superiority of the proposed method. At last, ablation studies are conducted on DUO dataset to validate the effectiveness and efficiency of each design in dynamic YOLO. Source code will be available at https://github.com/chenjie04/Dynamic-YOLO.

Bridging the gap: Active learning for efficient domain adaptation in object detection

In practical object detection computer vision applications, training commonly incorporates multiple data sources. Domain adaptation enhances the models’ capacity to generalize across different source and target domain data. Active learning, which serves as a bridge between unsupervised and supervised domain adaptation, is an area that has been underexplored for object detection. Our methods are specifically designed to tackle the challenges of selecting target domain images for annotation, complementing the existing source domain images. First, this paper formally describes active learning in domain adaptation for object detection. We propose three new selection strategies for active learning in domain adaptation for object detection. We utilize multiple machine learning models when selecting target domain data for annotation. In particular, our best-performing selection method utilizes a domain discriminator to predict the target domain likelihood of each sample. Additionally, we introduce an evaluation framework that evaluates per-box and per-image labeling efforts in the unique object detection application. Through extensive experiments we could show that our best-performing method could improve the active learning object detection performance over random selection by about 2.4% while requiring about 13% less annotated bounding boxes on the target domain.

Examining the nuances of Huizhou architecture and building decoration elements within the framework of rural development and urban aesthetics through the application of object detection and explicative analysis

Examining the nuances of Huizhou architecture and building decoration elements within the framework of rural development and urban aesthetics through the application of object detection and explicative analysis

YOLOv7-Bw: A Dense Small Object Efficient Detector Based on Remote Sensing Image

In recent years, deep learning techniques have been increasingly applied to the detection of remote sensing images. However, the substantial size variation and dense distribution of objects in these images present significant challenges to detection algorithms. Current methods often suffer from low efficiency, missed detections, and inaccurate bounding boxes. To address these issues, this paper presents an improved YOLO algorithm, YOLOv7-bw, designed for efficient remote sensing image detection, thereby advancing object detection applications in the remote sensing industry. YOLOv7-bw enhances the original SPPCSPC pooling pyramid network by incorporating a Bi-level Routing Attention module, which focuses on densely populated target areas to improve the network's feature extraction capabilities. Additionally, it introduces a dynamic non-monotonic WIoUv3 loss function to replace the original CIoU loss function. This substitution ensures that the loss function's gradient allocation strategy aligns more effectively with the current detection scenario, enhancing the network's focus on the detection object. Through comparative experiments on the DIOR remote sensing image dataset, we found that YOLOv7-bw achieved a high mAP@0.5 of 85.63% and a high mAP@0.5:0.95 of 65.93%, surpassing the previous results of 83.7% and 63.9% by approximately 1.93% and 2.03%, respectively. Moreover, compared with commonly used algorithms, YOLOv7-bw demonstrated superior performance, thereby validating the feasibility and enhanced applicability of our proposed algorithm for remote sensing image detection.

Object Detection using Convolutional Neural Network

During the last years, a noticeable growth is observed in the field of computer vision research. In computer vision, object detection is a task of classifying and localizing the objects in order to detect the same. The widely used object detection applications are human– computer interaction, video surveillance, satellite imagery, transport system, and activity recognition. In the wider family of deep learning architectures, convolutional neural network (CNN) made up with set of neural network layers is used for visual imagery. Deep CNN architectures exhibit impressive results for detection of objects in digital image. This paper represents a comprehensive review of the recent development in object detection using convolutional neural networks

MOD-YOLO: Multispectral object detection based on transformer dual-stream YOLO

Multispectral object detection can effectively improve the precision of object detection in low-visibility scenes, which increases the reliability and stability of the object detection application in the open environment. Cross-Modality Fusion Transformer (CFT) can effectively fuse different spectral information, but this method relies on large models and expensive computing resources. In this paper, we propose multispectral object detection dual-stream YOLO (MOD-YOLO), based on Cross Stage Partial CFT (CSP-CFT), to address the issue that prior studies need heavy inference calculations from the recurrent fusing of multispectral features. This network can divide the fused feature map into two parts, respectively for cross stage output and combined with the next stage feature, to achieve the correct speed/memory/precision balance. To further improve the accuracy, SIoU was selected as the loss function. Ultimately, extensive experiments on multiple publicly available datasets demonstrate that our model, which achieves the smallest model size and excellent performance, produces better tradeoffs between accuracy and model size than other popular models.

A roulette wheel-based pruning method to simplify cumbersome deep neural networks

Deep neural networks (DNNs) have been applied in many pattern recognition or object detection applications. DNNs generally consist of millions or even billions of parameters. These demanding computational storage and requirements impede deployments of DNNs in resource-limited devices, such as mobile devices, micro-controllers. Simplification techniques such as pruning have commonly been used to slim DNN sizes. Pruning approaches generally quantify the importance of each component such as network weight. Weight values or weight gradients in training are commonly used as the importance metric. Small weights are pruned and large weights are kept. However, small weights are possible to be connected with significant weights which have impact to DNN outputs. DNN accuracy can be degraded significantly after the pruning process. This paper proposes a roulette wheel-like pruning algorithm, in order to simplify a trained DNN while keeping the DNN accuracy. The proposed algorithm generates a branch of pruned DNNs which are generated by a roulette wheel operator. Similar to the roulette wheel selection in genetic algorithms, small weights are more likely to be pruned but they can be kept; large weights are more likely to be kept but they can be pruned. The slimmest DNN with the best accuracy is selected from the branch. The performance of the proposed pruning algorithm is evaluated by two deterministic datasets and four non-deterministic datasets. Experimental results show that the proposed pruning algorithm generates simpler DNNs while DNN accuracy can be kept, compared to several existing pruning approaches.

Smart Video Surveillance Using Deep Learning

Abstract: There is a lot of assessment happening in the business about video observation among them; the occupation of CCTV accounts has been blocked. CCTV cameras are put all around the spots for perception and security. In today's digital age, the increasing availability of data and advancements in computer vision have paved the way for numerous applications of object detection. This powerful technology has found its significance in various domains, including video surveillance and image retrieval systems. Object detection enables machines to identify and locate objects within images or video frames, providing valuable insights and aiding in decision-making processes. This article explores the applications, challenges, techniques, and future trends of object detection in the context of video surveillance and image retrieval systems. Object detection is a computer vision technique that involves identifying and localizing objects within images or video frames. Unlike image classification, which assigns a single label to an entire image, object detection goes a step further by detecting and delineating the individual objects present. It enables computers to comprehend visual data and interact with the world, much like humans do

Clinical Validation of Computer-Aided Diagnosis Software for Preventing Retained Surgical Sponges.

We previously reported the successful development of a computer-aided diagnosis (CAD) system for preventing retained surgical sponges with deep learning using training data, including composite and simulated radiographs. In this study, we evaluated the efficacy of the CAD system in a clinical setting. A total of 1,053 postoperative radiographs obtained from patients 20 years of age or older who underwent surgery were evaluated. We implemented a foreign object detection application software on the portable radiographic device used in the operating room to detect retained surgical sponges. The results of the CAD system diagnosis were prospectively collected. Among the 1,053 images, the CAD system detected possible retained surgical items in 150 images. Specificity was 85.8%, which is similar to the data obtained during the development of the software. The validation of a CAD system using deep learning in a clinical setting showed similar efficacy as during the development of the system. These results suggest that the CAD system can contribute to the establishment of a more effective protocol than the current standard practice for preventing the retention of surgical items.