Study on autonomous ship enhancement by optimizing the object detecting algorithms

Airplane and ship play a very important role in both civil life and military operations. It is a meaningful to detect airplane and ship around the world through remote sensing images. Target recognition algorithms based on deep learning technology are proven to be effective and gradually replacing traditional algorithms. This paper builds a target detection system based on NVIDIA TX2 development platform and YOLO V3 algorithm and focuses on both the ship and airplane targets. The training data comes from image fragments generated by satellites such as Jilin No. 1, DigitalGlobe, and Planet. The label of each target includes a bounding box and category information. The image processing method such as rotation and noise is added to increase the robustness of the trained YOLO V3 network for different sensors and atmosphere. The training of the input image takes about 3 days on an NVIDIA Titan X GPU. At test time, we partition testing images of arbitrary size into cutouts with a fixed size of 1 k × 1 k and run each cutout through our trained model to find ships and airplanes. The experimental results show that the F1-score values of the airport and the ship are 91.48% and 93.89%, respectively, and the detection speed of one cutout on NVIDIA TX2 development platform is about 0.56 s.

Synthetic Aperture Radar (SAR) is widely used in ship target detection because of its ability to operate under various weather conditions. However, some target images are difficult to obtain and label, resulting in a small sample size, which limits the development of target detection. Aiming at the problem of ship target detection in small sample SAR images, domain knowledge is used to revise and enhance the basic model in this paper. Firstly, a lightweight convolutional neural network model more suitable for small-sample SAR image ship target classification is proposed, involving target pixel and aspect ratio as domain knowledge to correct the classification results. Then, for ship target detection in SAR images, the acquisition method of scale class domain knowledge is improved. Moreover, texture-related domain knowledge based on the gray level co-occurrence matrix is extracted and used as classification features for the model to revise the model. Finally, Marine Targets Classification Dataset (MTCD) and Marine Targets Detection Dataset (MTDD) are briefly introduced and adjusted to meet the requirements of this research through screening and modification. The test results on balanced MTDD and small-sample datasets demonstrate the effectiveness of domain knowledge in improving network performance.

Object detection is a fundamental problem in computer vision, with applications spanning self-driving cars, surveillance systems, medical imaging, robotics, and smart cities. Among the myriad of algorithms developed for this task, the You Only Look Once (YOLO) family stands out for its ability to perform real-time and accurate object detection. This article provides a comprehensive analysis of the YOLO algorithm series, from YOLOv1 to YOLOv8, evaluating them across key performance metrics, including precision, recall, mean Average Precision (mAP), frames per second (FPS), and overall effectiveness. Unlike traditional two-stage detectors such as R-CNN, YOLO formulates object detection as a single regression problem: a single pass over the image simultaneously predicts bounding boxes and class probabilities. This end-to-end design enables YOLO to achieve high speed while maintaining a competitive accuracy-efficiency trade-off. We examine architectural innovations across YOLO versions, including batch normalization, anchor boxes, residual blocks, feature pyramid networks, and attention mechanisms, and discuss their impact on performance. Lightweight models (e.g., YOLOv5-Nano, YOLOv8-Small) are explored with a focus on their suitability for mobile and embedded systems, highlighting YOLO’s adaptability to resource-constrained environments. Challenges such as small object detection, occlusion, and domain-specific tuning are also addressed. This article serves as a practical guide for researchers, developers, and practitioners aiming to leverage YOLO for real-world object detection tasks.

Ship target detection on the sea surface is one of the most common scenes in the field of target detection. The accuracy of target detection plays an important role in the field of sea rescue and defense. However, complex infrared sea surface scenes, such as island shore, fish scale wave, sea surface bright band and other disturbances, bring great challenges to target detection. In this paper, we propose an improved YOLOV5 model by analyzing the characteristics of infrared image imaging and ship target on the sea surface. Aiming at the problem of information loss of small targets in the deep layer of the network, we redesigned the backbone network, which was composed of four Multi-scale residual blocks, and each block was connected by CBAM (Convolutional Block Attention Module) attention mechanism to improve information transmission and fusion between different feature layers. Aiming at the problem of complex sea surface scene interference, we use FPN and PAN (Feature Pyramid Network and Personal Area Network) mechanism to construct feature fusion network in neck, and add shallow feature fusion in FPN. Our model achieves good performance both in terms of latency and accuracy. In a self-collected sea surface scene dataset, multiple SOTAs for detection tasks are compared and the results demonstrate the superiority of the proposed method.

Due to the complex sea background and the different sizes and shapes of ships, the detection of ship targets has the problems of low detection rate, high false detection rate and high missed detection rate. To solve this problem, this paper proposes a multi-feature weighted fusion and PCA-SVM model for ship detection in remote sensing images. A weighted fusion method of Local Binary Pattern (LBP) and Histogram of Oriented Gradients (HOG) was used to extract shape and texture features simultaneously. Principal Component Analysis (PCA) was used to reduce the dimension of fused features to reduce the false detection rate caused by redundant noise interference. The feature pyramid was constructed to realize the feature fusion of different levels. In addition, a re-screening method based on color features or geometric features is proposed to further reduce the false detection rate. The accuracy and recall rate of the proposed detection algorithm in the DFH-MODIS dataset reach 97.9%and 79.8% respectively. Compared with the ship target detection algorithm based on HOG feature combined with SVM classifier, the F1-score is improved from 0.554 to 0.879. It effectively improves the detection performance of the ship target detection algorithm in optical remote sensing images, and has better effectiveness and robustness.

Object detection in aerial images plays a crucial role across diverse domains such as agriculture, environmental monitoring, and security. Aerial images present several challenges, including dense small objects, intricate backgrounds, and occlusions, necessitating robust detection algorithms. This paper addresses the critical need for accurate and efficient object detection in aerial images using a Transformer-based approach enhanced with specialized methodologies, termed DFS-DETR. The core framework leverages RT-DETR-R18, integrating the Cross Stage Partial Reparam Dilation-wise Residual Module (CSP-RDRM) to optimize feature extraction. Additionally, the introduction of the Detail-Sensitive Pyramid Network (DSPN) enhances sensitivity to local features, complemented by the Dynamic Scale Sequence Feature-Fusion Module (DSSFFM) for comprehensive multi-scale information integration. Moreover, Multi-Attention Add (MAA) is utilized to refine feature processing, which enhances the model’s capacity for understanding and representation by integrating various attention mechanisms. To improve bounding box regression, the model employs MPDIoU with normalized Wasserstein distance, which accelerates convergence. Evaluation across the VisDrone2019, AI-TOD, and NWPU VHR-10 datasets demonstrates significant improvements in the mean average precision (mAP) values: 24.1%, 24.0%, and 65.0%, respectively, surpassing RT-DETR-R18 by 2.3%, 4.8%, and 7.0%, respectively. Furthermore, the proposed method achieves real-time inference speeds. This approach can be deployed on drones to perform real-time ground detection.

To further improve the speed and accuracy of object detection, especially small targets and occluded objects, a novel and efficient detector named YOLO-ACN is presented. The detector model is inspired by the high detection accuracy and speed of YOLOv3, and it is improved by the addition of an attention mechanism, a CIoU (complete intersection over union) loss function, Soft-NMS (non-maximum suppression), and depthwise separable convolution. First, the attention mechanism is introduced in the channel and spatial dimensions in each residual block to focus on small targets. Second, CIoU loss is adopted to achieve accurate bounding box (BBox) regression. Besides, to filter out a more accurate BBox and avoid deleting occluded objects in dense images, the CIoU is applied in the Soft-NMS, and the Gaussian model in the Soft-NMS is employed to suppress the surrounding BBox. Third, to significantly reduce the parameters and improve the detection speed, standard convolution is replaced by depthwise separable convolution, and hard-swish activation function is utilized in deeper layers. On the MS COCO dataset and infrared pedestrian dataset KAIST, the quantitative experimental results show that compared with other state-of-the-art models, the proposed YOLO-ACN has high accuracy and speed in detecting small targets and occluded objects. YOLO-ACN reaches a mAP50 (mean average precision) of 53.8% and an APs (average precision for small objects) of 18.2% at a real-time speed of 22 ms on the MS COCO dataset, and the mAP for a single class on the KAIST dataset even reaches over 80% on an NVIDIA Tesla K40.

Object detection and tracking is one of the key applications of wireless sensor networks (WSNs). The key issues associated with this application include network lifetime, object detection and localization accuracy. To ensure the high quality of the service, there should be a trade-off between energy efficiency and detection accuracy, which is challenging in a resource-constrained WSN. Most researchers have enhanced the application lifetime while achieving target detection accuracy at the cost of high node density. They neither considered the system cost nor the object localization accuracy. Some researchers focused on object detection accuracy while achieving energy efficiency by limiting the detection to a predefined target trajectory. In particular, some researchers only focused on node clustering and node scheduling for energy efficiency. In this study, we proposed a mobile object detection and tracking framework named the Energy Efficient Object Detection and Tracking Framework (EEODTF) for heterogeneous WSNs, which minimizes energy consumption during tracking while not affecting the object detection and localization accuracy. It focuses on achieving energy efficiency via node optimization, mobile node trajectory optimization, node clustering, data reporting optimization and detection optimization. We compared the performance of the EEODTF with the Energy Efficient Tracking and Localization of Object (EETLO) model and the Particle-Swarm-Optimization-based Energy Efficient Target Tracking Model (PSOEETTM). It was found that the EEODTF is more energy efficient than the EETLO and PSOEETTM models.

Accurate identification of road object is crucial for intelligent traffic systems. However, due to the complexity of road traffic scenarios, developing efficient and accurate road object detection methods has been a challenging task. In this study, a new improved method for road object detection is proposed, named Enhanced YOLOv5 algorithm and Deep Schedule Object Detection (DSOD) algorithm. Real traffic scenes from the BDD100k dataset are used for training and testing the object detection model. The dataset consists of 9 different types of road objects in various traffic scenarios. The Mosaic data augmentation algorithm is applied to merge images in the dataset. Mean Average Precision (mAP), Precision (P), and Recall (R) are used as evaluation metrics to compare the enhanced YOLOv5 model with the most common models. Experimental results demonstrate that the DSOD algorithm achieves success in intelligent traffic object detection, significantly improving the accuracy and robustness of road object recognition. Additionally, the developed model shows significant performance improvement in accurately identifying objects in complex traffic scenes. These results suggest that the DSOD algorithm is a promising choice for intelligent road recognition and can easily adapt to different traffic scenarios. Furthermore, employing cloud computing for real-time detection meets the requirements of intelligent cooperative vehicles and enhances their visual perception capabilities.

With the rapid development of urbanization and global climate warming, complex environments and adverse weather conditions pose significant challenges to the accuracy of object detection and driving safety in autonomous driving systems. Addressing the challenges of severe occlusion and noise interference in target detection within complex foggy scenes and considering the YOLOv5 model’s small size and fast processing capabilities, which meet the real-time processing demands in complex environments, it is particularly suited for resource-constrained vehicular systems. Consequently, this paper introduces the YOLOv5-RCBiW model tailored for vehicular vision perception aimed at enhancing feature extraction and recognition. Initially, the Receptive Field Block (RFB) is integrated with the Coordinate Attention (CA) mechanism to form the RFCA module, which emphasizes the importance of different features and optimizes receptive field spatial features. Furthermore, the Re-BiFPN module is constructed to enhance feature perception accuracy through bidirectional cross-scale connections and feature fusion, while the detection head at the P5 layer is replaced to improve recognition capabilities. Finally, a gradient gain loss function is introduced to reduce feature information loss and prevent model performance degradation, ensuring robustness and accuracy in complex environments. The comparative experimental results on the RTTS and Foggy Driving datasets indicate that the YOLOv5-RCBiW model significantly outperforms existing models in object detection accuracy under foggy and complex scenes. Additionally, in-vehicle experiments validate the model’s effectiveness and real-time performance in challenging environments.

Vehicle transportation of hazardous chemicals is one of the important mobile hazards in modern logistics, and its unsafe factors bring serious threats to people's lives, property and environmental safety. Although the current object detection algorithm has certain applications in the detection of hazardous chemical vehicles, due to the complexity of the transportation environment, the small size and low resolution of the vehicle target etc., object detection becomes more difficult in the face of a complex background. In order to solve these problems, the authors propose an improved algorithm based on YOLOv5 to enhance the detection accuracy and efficiency of hazardous chemical vehicles. Firstly, in order to better capture the details and semantic information of hazardous chemical vehicles, the algorithm solves the problem of mismatch between the receptive field of the detector and the target object by introducing the receptive field expansion block into the backbone network, so as to improve the ability of the model to capture the detailed information of hazardous chemical vehicles. Secondly, in order to improve the ability of the model to express the characteristics of hazardous chemical vehicles, the authors introduce a separable attention mechanism in the multi‐scale target detection stage, and enhances the prediction ability of the model by combining the object detection head and attention mechanism coherently in the feature layer of scale perception, the spatial location of spatial perception and the output channel of task perception. Experimental results show that the improved model significantly surpasses the baseline model in terms of accuracy and achieves more accurate object detection. At the same time, the model also has a certain improvement in inference speed and achieves faster inference ability.

With the rapid development of optical remote sensing, it is urgent to find a reliable target detection method. Compared with traditional detection algorithms, a convolutional neural network has attracted considerable attention owing to its efficiency and high transitivity. However, different from general images, remote sensing images contain complex background information and dense small targets with changeable directions that make detection very challenging. To solve these problems and provide a stable and high-performance detection method, a rotated saliency fusion object detection (RSD) model based on “you only look once” (YOLO)v4 is established. First, salient image fusion technology is used to magnify target information. Second, the angle variable and rotated non-maximal suppression is introduced to improve the accuracy of rotated object detection by including the detection of dense objects. Third, the network structure is enhanced to improve the performance of small-target detection. Finally, the k-means algorithm and data enhancement are introduced to increase the robustness of the model. Extensive experiments demonstrate the superiority of the proposed model in detection speed and accuracy. The mean average precision of the proposed RSD model reaches 97.32% for the remote sensing images in a harbor area with an average detection speed of 13.41 s − 1.

Accurate and reliable target detection is a crucial requirement of high-frequency surface wave radar for effective maritime surveillance. However, existing methods based on single-frame radar images primarily focus on static target features, limiting their ability to capture dynamic ship behaviors across multiple frames. In this study, a two-stage multiframe target detection framework (MFTDF) is proposed to address this issue in ship-target detection from range-Doppler (RD) images. The framework consists of two stages: a space-to-depth YOLOv8 network in Stage-1 and a multiframe vision transformer network in Stage-2. First, Stage-1 aims to extract regions of interest (ROI) from the current and several preceding RD images. Thereafter, target association and pattern selection are applied to collect multiframe image patch sequences for each ROI result. Finally, Stage-2 focuses on further discriminating the input image patches to obtain refined target detection results. Moreover, the dataset for training and validating the two-stage network is automatically generated based on the automatic identification system ship data and constant false alarm rate detection results to identify all true and visible targets, ensuring the dataset's credibility. Experiments using the measured data show that the proposed MFTDF achieves a considerably improved precision rate while maintaining an average improvement of 12.3% in the recall rate. These results confirm that MFTDF delivers superior detection accuracy and efficiency, offering a robust solution for maritime target detection in complex scenarios.

Ship targets are important carriers of maritime behavior, and ship target detection has important economic and military value. Owing to its excellent characteristics, synthetic aperture radar (SAR) is widely used to detect sea targets. SAR ship detection has always been a popular field of research. The excellent performance of deep neural networks in computer vision has facilitated their application to SAR image detection. However, several problems remain in SAR ship detection at this stage. First, SAR ship targets differ significantly in size, and how to deal with multi-scale ship detection has, therefore, been a difficult research topic. Second, the size of a SAR ship target is small in low-resolution images, and thus, dealing with small size target detection is also difficult. Therefore, we propose a dense connection DetNet (DDNet), which is specifically used for SAR ship detection. The network consists of two parts: backbone sub-network and prediction sub-network. In the backbone subnetwork, we refer to DetNet, and use a stacked convolution layer instead of large downsampling to make it more suitable for small ship detection. In the prediction sub-network, we use a dense connection to fuse features of different scales, allowing it to better deal with multi-scale ship detection. The feature reuse strategy is used to improve the parameter efficiency. The experiments on SAR ship datasets show that our method performs better than other methods in small-size and complex background ship detection.

Affected by wave, illumination intensity, water mist, and other factors, the traditional target detection algorithm is difficult to accurately and robustly identify water surface targets. To solve this problem, this paper proposes water surface targets detection method of USV based on deep learning. First of all, the ASFF was introduced to improve the basic YOLOv5s model, and then the improved YOLOv5s model was trained and tested with self-made surface data set. Experimental results show that the improved model has stronger feature extraction ability, the model detection accuracy is improved by 9.4%, and the actual detection speed reaches 138 FPS.