Object Detection Applications Research Articles

Deep learning-based object detection in maritime unmanned aerial vehicle imagery: Review and experimental comparisons

With the advancement of maritime unmanned aerial vehicles (UAVs) and deep learning technologies, the application of UAV-based object detection has become increasingly significant in the fields of maritime industry and ocean engineering. Endowed with intelligent sensing capabilities, the maritime UAVs enable effective and efficient maritime surveillance. To further promote the development of maritime UAV-based object detection, this paper provides a comprehensive review of challenges, relative methods, and UAV aerial datasets. Specifically, in this work, we first briefly summarize four challenges for object detection on maritime UAVs, i.e., object feature diversity, device limitation, maritime environment variability, and dataset scarcity. We then focus on computational methods to improve maritime UAV-based object detection performance in terms of scale-aware, small object detection, view-aware, rotated object detection, lightweight methods, and others. Next, we review the UAV aerial image/video datasets and propose a maritime UAV aerial dataset named MS2ship for ship detection. Furthermore, we conduct a series of experiments to present the performance evaluation and robustness analysis of object detection methods on maritime datasets. Eventually, we give the discussion and outlook on future works for maritime UAV-based object detection. The MS2ship dataset is available at https://github.com/zcj234/MS2ship.

Battle Models: Inception ResNet vs. Extreme Inception for Marine Fish Object Detection

Within the domain of deep learning applied to computer vision, there exists a significant emphasis on the competition between two prominent models, namely Inception ResNet and Xception, particularly in the field of marine fish object detection. The present study conducted a comparative analysis of two advanced neural network architectures in order to assess their efficacy in the identification and localization of marine fish species in underwater images. The two models underwent a rigorous evaluation, utilizing their capabilities in feature extraction. The findings indicate a complex performance landscape, wherein Inception ResNet exhibits remarkable accuracy in identifying marine fish objects, while Xception demonstrates superior computational efficiency. The present study elucidates the inherent trade-off between precision and computational expenditure, offering valuable perspectives on the pragmatic ramifications of choosing one model over another. Furthermore, this research underscores the significance of carefully choosing a suitable model that aligns with the particular requirements of object detection applications in the context of marine fish. This study endeavors to guide professionals and scholars in marine biology and computer vision, enabling them to make well-informed choices when utilizing deep learning techniques to detect maritime fish objects in underwater settings. The research specifically focuses on the comparison between Inception ResNet and Xception models.

Performance Comparison of Sea Cucumber Detection by the Yolov5 and DETR Approach

Sea cucumber detection represents an important step in underwater environmental perception, which is an indispensable part of the intelligent subsea fishing system. However, water turbidity decreases the clarity of underwater images, presenting a challenge to vision-based underwater target detection. Therefore, accurate, real-time, and lightweight detection models are required. First of all, the development of subsea target detection is summarized in this present work. Object detection methods based on deep learning including YOLOv5 and DETR, which are, respectively, examples of one-stage and anchor-free object detection approaches, have been increasingly applied in underwater detection scenarios. Based on the state-of-the-art underwater sea cucumber detection methods and aiming to provide a reference for practical subsea identification, adjacent and overlapping sea cucumber detection based on YOLOv5 and DETR are investigated and compared in detail. For each approach, the detection experiment is carried out on the derived dataset, which consists of a wide variety of sea cucumber sample images. Experiments demonstrate that YOLOv5 surpasses DETR in low computing consumption and high precision, particularly in the detection of small and dense features. Nevertheless, DETR exhibits rapid development and holds promising prospects in underwater object detection applications, owing to its relatively simple architecture and ingenious attention mechanism.

Analysis of convolutional neural networks and its application in object detection

With the development of times and technology, artificial intelligence-based neural network algorithms have been widely used in scientific research and life. Among them, Convolutional neural network (CNN) is the most classic and the most representative. CNN have been widely used in images classification in recent years. This article focuses on the basic information of the convolutional neural networks and its applications especially in object detection. Also, the advantages of the CNN and the possible future improvements will be told in this article. In the end of this article some research of experiments will be shown in order to prove the accuracy of convolutional neural networks in detecting objects and visualizing results. This article gathers basic knowledge of convolutional neural networks. As a result, it may help new starters to get to know the CNN and then make contributions to the development of deep learning especially the CNN according to the advantages and future improvements mentioned in the article.

Survey of object detection in terahertz images

In recent years, object detection algorithms have undergone a further development and improvement, resulting in a wider range of application scenarios. As one of the most fundamental and challenging issues in the field of computer vision, the application of object detection in the field of security has also received considerable attention. Terahertz (THz) imaging which is widely used in this area because of the ability to detect hidden objects, as a type of electromagnetic wave imaging with poor imaging performance and low resolution, traditional target detection methods cannot achieve high robustness and effectiveness simultaneously. However, anyway in this possible application scenario, many possible ideas and algorithms have been proposed. This paper analyzes the possibility of applying different object detection methods to terahertz images and analyzes the existing problems in order to give the learner some basic idea and future direction. Several detectors are covered, including the traditional object detection methods and the algorithms based on Convolutional Neural Network(CNN) framework.

Review of Adversarial Attacks in Object Detection

Object detection, a fundamental element of computer vision and artificial intelligence, has experienced considerable advancements through the incorporation of deep learning-based techniques. Yet, despite the impressive strides in both accuracy and efficiency, object detection algorithms harbor inherent vulnerabilities to adversarial attacks. These well-crafted disruptions pose significant risks, especially considering the broad application of object detection across an array of safety-critical sectors such as autonomous transportation, medical imaging, and security systems. This comprehensive paper offers a thorough review of adversarial attacks against object detection systems, dissecting the methods employed, and scrutinizing the implications of their exploits. It dives deep into the mechanics and consequences of both white-box and black-box attacks on prevalent object detection networks, including but not limited to Faster R-CNN, YOLO, and SSD. Furthermore, this paper underscores an assortment of defense strategies developed to mitigate the effects of adversarial attacks. These include adversarial training, gradient masking, input transformations, and randomized defenses. While these strategies hold promise, it is acknowledged that they have their limitations and do not offer a universal solution against all adversarial attacks. As such, this paper underscores the urgent necessity for robust defense mechanisms and stimulates further discourse and investigation into developing truly resilient object detection systems, capable of withstanding the growing threat of adversarial attacks.

Object Detection and Distance Measurement Algorithm for Collision Avoidance of Precast Concrete Installation during Crane Lifting Process

In the construction industry, the process of carrying heavy loads from one location to another by means of a crane is inevitable. This reliance on cranes to carry heavy loads is more obvious when it comes to high-rise building construction. Depending on the conditions and requirements on-site, various types of construction lifting equipment (i.e., cranes) are being used. As off-site construction (OSC) is gaining more traction recently, cranes are becoming more important throughout the construction project as precast concrete (PC) members are major components of OSC calling for lifting work. As a result of the increased use of cranes on construction sites, concerns about construction safety as well as the effectiveness of existing load collision prevention systems are attracting more attention from various parties involved. Besides the inherent risks associated with heavy load lifting, the unpredictable movement of on-site workers around the crane operation area, along with the presence of blind spots that obstruct the crane operator’s field-of-view (FOV), further increase the accident probability during crane operation. As such, the need for a more reliable and improved collision avoidance system that prevents lifted loads from hitting other structures and workers is paramount. This study introduces the application of deep learning-based object detection and distance measurement sensors integrated in a complementary way to achieve the stated need. Specifically, the object detection technique was used with the application of an Internet Protocol (IP) camera to detect the workers within the crane operation radius, whereas ultrasonic sensors were used to measure the distance of surrounding obstacles. Both applications were designed to work concurrently so as to prevent potential collisions during crane lifting operations. The field testing and evaluation of the integrated system showed promising results.

3D lidar SLAM-based systems in object detection and navigation applications

ABSTRACT This paper considered an object detection system based on 3D LiDAR Sensor and Simultaneous Localization and Mapping (SLAM) to complete the navigation applications of mobile robots. A 3D-based SLAM with lightweight and ground-optimized Lidar odometry and mapping (LeGO-LOAM) appropriately generated the environmental maps. SLAM is a tool used to obtain information from the environment, allowing mobile robots to know their location. Indoor environment data is immedicably created while SLAM is processing the information. The dynamic object detection algorithm depends on the available information to realize the external morphology and circle the bounding box of moving objects. Therefore, a wheeled mobile robot (WMR) was employed to dynamically trace the object’s movement direction. Finally, This study found that the quantum genetic algorithm (QGA) is more efficient in generating a shorter path than the particle swarm optimization, and a dynamic window approach (DWA) is immediately detected as a dynamic obstacle. Therefore, WMR obtains enough object, obstacle, and routing information to effectively and safely reach the destination through the Move_base software package in Robot Operating System.

A Systematic Stereo Camera Calibration Strategy: Leveraging Latin Hypercube Sampling and 2k Full-Factorial Design of Experiment Methods.

This research aimed to optimize the camera calibration process by identifying the optimal distance and angle for capturing checkered board images, with a specific focus on understanding the factors that influence the reprojection error (ϵRP). The objective was to improve calibration efficiency by exploring the impacts of distance and orientation factors and the feasibility of independently manipulating these factors. The study employed Zhang's camera calibration method, along with the 2k full-factorial analysis method and the Latin Hypercube Sampling (LHS) method, to identify the optimal calibration parameters. Three calibration methods were devised: calibration with distance factors (D, H, V), orientation factors (R, P, Y), and the combined two influential factors from both sets of factors. The calibration study was carried out with three different stereo cameras. The results indicate that D is the most influential factor, while H and V are nearly equally influential for method A; P and R are the two most influential orientation factors for method B. Compared to Zhang's method alone, on average, methods A, B, and C reduce ϵRP by 25%, 24%, and 34%, respectively. However, method C requires about 10% more calibration images than methods A and B combined. For applications where lower value of ϵRP is required, method C is recommended. This study provides valuable insights into the factors affecting ϵRP in calibration processes. The proposed methods can be used to improve the calibration accuracy for stereo cameras for the applications in object detection and ranging. The findings expand our understanding of camera calibration, particularly the influence of distance and orientation factors, making significant contributions to camera calibration procedures.

Deep Convolutional Neural Network Algorithm for Prediction of the Mechanical Properties of Friction Stir Welded Copper Joints from its Microstructures

Convolutional Neural Network (CNN) is a special type of Artificial Neural Network which takes input in the form of an image. Like Artificial Neural Network they consist of weights that are estimated during training, neurons (activation functions), and an objective (loss function). CNN is finding various applications in image recognition, semantic segmentation, object detection, and localization. The present work deals with the prediction of the welding efficiency of the Friction Stir Welded joints on the basis of microstructure images by carrying out training on 3000 microstructure images and further testing on 300 microstructure images. The obtained results showed an accuracy of 80 % on the validation dataset.

Automatic labeling to overcome the limitations of deep learning in applications with insufficient training data: A case study on fruit detection in pear orchards

Accurate counting of pears in orchard environments is essential for crop management. However, due to the time and labor cost of manual counting, farmers rely on sampling a few trees and extending the count to the entire orchard, which overlooks the spatial variability of fruit load. This study proposes automatic labeling in case of insufficient manually labeled data to accelerate the adaptation of on-tree fruit counting solutions based on deep learning. Since the amount of training data needed can be one of the main limitations to deploying a robust solution in the field, here we present an approach to minimize human intervention in data labeling. We describe, first, how the latest advances in self-supervised learning and transformers can be used to create an automatic labeling procedure, and second, how the automatically labeled dataset can be exploited despite its noise. For the dataset creation using automatic labeling, we modified a single object discovery algorithm to localize multiple object instances, and we trained an image classifier to include only objects from our target class: pears. The automatically generated labels are regarded as ‘noisy’ because the quality of the annotations is lower than manually annotated data. We then pre-trained an object detection model with the automatically labeled data (121,038 annotations) and fine-tuned it with insufficient manually labeled data (500, 1000, and 2500 annotations). This allowed us to leverage knowledge from both datasets despite the quality and size restrictions. The performance comparison of object detection models trained this way to those trained only with insufficient manually labeled data showed average precision gains of 37.0, 21.4, and 1.0%. In addition, on unseen images taken with the same camera setup, the average precision gains were 35.9, 17.6, and 1.8%. Furthermore, gains were 38.5, 25.1, and 1.2% on unseen images taken with a different camera setup. These results suggest the substantial value of automatic labeling when having insufficient training data for fruit detection in pear orchards. Moreover, other object detection applications can replicate the presented methods and experiments, especially when resource constraints or rapid validation prevent extensive manual labeling.

Faster R-CNN and YOLOv3: a general analysis between popular object detection networks

Computer vision is developing really fast in recent years, and object detection is now one of the hottest topics. faster region-based convolutional neural network (faster R-CNN) and You Only Look Once (YOLOv3) are two popular modules to solve object detection problems, but they are implemented in different ways and thus have different performance in practice. This article is going to introduce these two modern modules and make a few experiments to compare the performance of each module with various datasets. This analysis will focus on speed, accuracy and the performances in different situations. Two specific studies will be mentioned as two typical examples of object detection application: face mask detection and greenhouse detection from satellite images. At last, the article will draw out a conclusion, make some suggestions for the choice of faster R-CNN and YOLOv3 and make a prospect for the future. It turns out that generally faster R-CNN fits tasks that require high accuracy better and YOLOv3 can realize real-time detection tasks.

Computer Vision Application in Object Detection and Tracking for Aerial Surveillance

Computer Vision Application in Object Detection and Tracking for Aerial Surveillance

EFECL: Feature encoding enhancement with contrastive learning for indoor 3D object detection

Good proposal initials are critical for 3D object detection applications. However, due to the significant geometry variation of indoor scenes, incomplete and noisy proposals are inevitable in most cases. Mining feature information among these “bad” proposals may mislead the detection. Contrastive learning provides a feasible way for representing proposals, which can align complete and incomplete/noisy proposals in feature space. The aligned feature space can help us build robust 3D representation even if bad proposals are given. Therefore, we devise a new contrast learning framework for indoor 3D object detection, called EFECL, that learns robust 3D representations by contrastive learning of proposals on two different levels. Specifically, we optimize both instance-level and category-level contrasts to align features by capturing instance-specific characteristics and semantic-aware common patterns. Furthermore, we propose an enhanced feature aggregation module to extract more general and informative features for contrastive learning. Evaluations on ScanNet V2 and SUN RGB-D benchmarks demonstrate the generalizability and effectiveness of our method, and our method can achieve 12.3% and 7.3% improvements on both datasets over the benchmark alternatives. The code and models are publicly available at https://github.com/YaraDuan/EFECL.

A Symbol Recognition System for Single-Line Diagrams Developed Using a Deep-Learning Approach

In numerous electrical power distribution systems and other engineering contexts, single-line diagrams (SLDs) are frequently used. The importance of digitizing these images is growing. This is primarily because better engineering practices are required in areas such as equipment maintenance, asset management, safety, and others. Processing and analyzing these drawings, however, is a difficult job. With enough annotated training data, deep neural networks perform better in many object detection applications. Based on deep-learning techniques, a dataset can be used to assess the overall quality of a visual system. Unfortunately, there are no such datasets for single-line diagrams available to the general research community. To augment real image datasets, generative adversarial networks (GANs) can be used to create a variety of more realistic training images. The goal of this study was to explain how deep-convolutional-GAN- (DCGAN) and least-squares-GAN- (LSGAN) generated images are evaluated for quality. In order to improve the datasets and confirm the effectiveness of synthetic datasets, our work blended synthetic images with actual images. Additionally, we added synthetic images to the original picture collection to prepare an augmented dataset for symbol detection. In this scenario, we employed You Look Only Once (YOLO) V5, one of the versions of YOLO. The recognition performance was improved, reaching an accuracy of 95% with YOLO V5, after combining the actual images with the synthetic images created by the DCGAN and LSGAN. By incorporating synthetic samples into the dataset, the overall quality of the training data was improved, and the learning process for the model became simpler. Furthermore, the proposed method significantly improved symbol detection in SLDs, according to the findings of the experiments.

Progress in Object Detection: An In-Depth Analysis of Methods and Use Cases

Object detection, a fundamental task in computer vision, involves identifying and localizing objects within images or videos. This paper provides a comprehensive review of traditional and deep learning-based object detection techniques and their applications, challenges, and future directions. We first discuss traditional object detection methods, which rely on handcrafted features and classical machine learning algorithms. We then explore the advancements brought by deep learning, including convolutional neural networks (CNNs) and transformer-based architectures, which have significantly improved the accuracy and efficiency of object detection tasks. A thorough comparison and evaluation of different object detection techniques are presented, considering performance metrics, speed, and robustness to object size, orientation, and occlusion variations. We also examine the diverse applications of object detection across various domains, such as robotics, autonomous vehicles, surveillance, medical imaging, and augmented reality. We outline open challenges and future research directions, emphasizing the need to combine object detection with other tasks, develop few-shot and zero-shot learning approaches, and address issues related to fairness, accountability, and transparency. This paper aims to comprehensively review the most prominent object detection techniques, their evolution, and their applications in diverse domains. We discussed traditional methods and recent deep learning-based approaches, emphasizing their strengths and limitations.

Energy-Efficient Approximate Edge Inference Systems

The rapid proliferation of the Internet of Things and the dramatic resurgence of artificial intelligence based application workloads have led to immense interest in performing inference on energy-constrained edge devices. Approximate computing (a design paradigm that trades off a small degradation in application quality for disproportionate energy savings) is a promising technique to enable energy-efficient inference at the edge. This article introduces the concept of an approximate edge inference system ( AxIS ) and proposes a systematic methodology to perform joint approximations between different subsystems in a deep neural network (DNN)-based edge inference system, leading to significant energy benefits compared to approximating individual subsystems in isolation. We use a smart camera system that executes various DNN-based image classification and object detection applications to illustrate how the sensor, memory, compute, and communication subsystems can all be approximated synergistically. We demonstrate our proposed methodology using two variants of a smart camera system: (a) Cam Edge , where the DNN is executed locally on the edge device, and (b) Cam Cloud , where the edge device sends the captured image to a remote cloud server that executes the DNN. We have prototyped such an approximate inference system using an Intel Stratix IV GX-based Terasic TR4-230 FPGA development board. Experimental results obtained using six large DNNs and four compact DNNs running image classification applications demonstrate significant energy savings (≈ 1.6× -4.7× for large DNNs and ≈ 1.5× -3.6× for small DNNs), for minimal (<1%) loss in application-level quality. Furthermore, results using four object detection DNNs exhibit energy savings of ≈ 1.5× -5.2× for similar quality loss. Compared to approximating a single subsystem in isolation, AxIS achieves 1.05× -3.25× gains in energy savings for image classification and 1.35× -4.2× gains for object detection on average, for minimal (<1%) application-level quality loss.

Quantifying noise limitations of neural network segmentations in high-resolution transmission electron microscopy

Motivated by the need for low electron dose transmission electron microscopy imaging, we report the optimal frame dose (i.e.e−/Å2) range for object detection and segmentation tasks with neural networks. The MSD-net architecture shows promising abilities over the industry standard U-net architecture in generalising to frame doses below the range included in the training set, for both simulated and experimental images. It also presents a heightened ability to learn from lower dose images. The MSD-net displays mild visibility of a Au nanoparticle at 20–30 e−/Å2, and converges at 200 e−/Å2 where a full segmentation of the nanoparticle is achieved. Between 30 and 200 e−/Å2 object detection applications are still possible. This work also highlights the importance of modelling the modulation transfer function when training with simulated images for applications on images acquired with scintillator based detectors such as the Gatan Oneview camera. A parametric form of the modulation transfer function is applied with varying ranges of parameters, and the effects on low electron dose segmentation is presented.

A Novel L-CLAHE-Based Intensification Filter for Enhancement of Underwater Images and Pipeline Tracking

The high contrast degraded underwater images often demand extensive pre-processing to attain reliable information for object detection and pipeline tracking. The prototype system developed employs CLAHE, a variant of adaptive histogram equalization, with an adaptive clip limit based on entropy. To yield an enhanced image, CLAHE is applied to the L* channel alone, resulting in improved L- CLAHE and better results. Validation of the improved L-CLAHE method is carried out using real-life underwater images captured using DTG2 Pro ROV from Kochi backwaters based on PSNR, SSIM, EPI, and PFOM quality measures and outperforms the conventional histogram equalization and adaptive histogram equalization techniques. These images are subjected to intensification and filtering for further enhancement as well as object detection and pipeline tracking applications. The proposed algorithm is compared with CLAHE, MSRCR, and L-CIF methods. The improved L-CIF algorithm has a high entropy value compared to the existing algorithms while the EME metric has been evaluated close to unity indicating the improved performance of the fast-guided filter. The method developed also helps in improving the average pixel intensity of the input images.

Real-Time Automatic Assisted Detection of Uterine Fibroid in Ultrasound Images Using a Deep Learning Detector.

Uterine smooth muscle hyperplasia causes a tumor called a uterine fibroid. With an incidence of up to 30%, it is one of the most prevalent tumors in women and has the third highest prevalence of all gynecological illnesses. Although uterine fibroids are usually not accompanied by symptoms, there are physical effects, such as impairment of the ability to conceive. To reduce morbidity, early detection and treatment are crucial. Ultrasound imaging is a common method used for pre-operative guidance and interventional therapy. Many applications of object detection are performing well with the advancement of deep learning in the field of medical image analysis. To ensure accuracy, computer-assisted detection can further solve the subjective problem generated by different doctors when they read images. Our research provides an improved YOLOv3 that combines the properties of EfficientNet and YOLOv3, which use a convolutional neural network to extract features, to detect uterine fibroid ultrasound images. Our approach attained an F1 score of 95% and an average precision of 98.38% and reached a detection speed of 0.28 s per image. We reviewed and analyzed several detection techniques and identified potential future research hotpots. This technique offers enough supplementary diagnostic tools for amateur or expert ultrasonologists and sets a solid foundation for future medical care and surgical excision.