Abstract Deep learning has made significant progress in the field of low‐light image enhancement. However, challenges remain, such as the substantial parameter consumption required for effective image enhancement. Inspired by multi‐scale geometric transformations in image detail enhancement, a novel model called the fixed‐directional filters network is proposed. Fixed‐directional filters network takes the original image as input and employs multiple branches for parallel processing. One branch uses conventional convolutional layers to extract features from the original image, while the other branches apply non‐linear mapping layers based on wavelet transforms. These wavelet transform branches capture the multi‐scale information of the image by combining different directions and convolutional kernels and utilize a trainable custom gamma mapping layer for non‐linear modulation to enhance specific regions of the image. The feature maps processed by each branch are merged through concatenation operations and then passed through convolutional layers to output the enhanced image. Using trainable mapping functions alone to enhance details significantly reduces the reliance on convolutional layers, effectively lowering the model's parameter count to only 13k parameters. Additionally, experiments demonstrate that fixed‐directional filters network significantly improves image quality, particularly in capturing image details and enhancing image contrast.

Abstract The non‐local imaging characteristics of computational ghost imaging (CGI) make it widely used in the field of information security. However, the information processing method based on traditional CGI has some problems in the process of multi‐image encryption, such as low efficiency and unsatisfactory reconstruction quality. Therefore, this article proposes to use wavelet transform and singular value decomposition technology to help computational ghost imaging to efficiently complete the coding and encryption of multiple images, and use alternating direction multiplier method (ADMM) in the process of image restoration to carry out high‐quality reconstruction of the original encrypted image. The results show that the scheme has good reconstruction effect, strong security and robustness even at a low sampling rate. It provides some reference for digital watermarking technology and cryptography.

Abstract Apple leaf lesions present a challenge for their detection and recognition because of their wide variety of species, morphologies, uneven sizes, and complex backgrounds. This paper proposes an improved multi‐scale YOLOv8 for apple leaf dense lesion detection and recognition. In the proposed YOLOv8, an improved C2f‐RFEM module is constructed in the backbone network to improve the feature extraction of disease object. A new neck network is designed by using C2f‐DCN and C2f‐DCN‐EMA module, which are established with deformable convolutions and efficient multi‐scale attention module with cross‐spatial learning attention mechanism. Moreover, a large‐scale detection head is introduced for increasing the resolution of the small lesion targets, so as to further improve the detection ability for multi‐scale diseases. Finally, the improved YOLOv8 is tested on the common objects in context (COCO) database with 80 kinds of objectives and an apple leaf disease database with 8 kinds of diseases. Compared to the baseline YOLOv8 model, the proposed improved YOLOv8 increases the mAP0.5 by 3%, and decreases the floating‐point operations per second (FLOPs) by 0.3G on the COCO database. For the apple leaf disease database, the improved YOLOv8 outperforms in terms of mAP and FLOPs compared to other models, for parameters and model size, it is ranked second and third, respectively. Experimental results show that the improved YOLOv8 has better adaptability to multi‐scale dense distribution of apple leaf disease spots with complex scenarios.

Abstract Medical image registration is essential and a key step in many advanced medical image tasks. In recent years, medical image registration has been applied to many clinical diagnoses, but large deformation registration is still a challenge. Deep learning‐based methods typically have higher accuracy but do not involve spatial transformation, which ignores some desirable properties, including topology preservation and the invertibility of transformation, for medical imaging studies. On the other hand, diffeomorphic registration methods achieve a differentiable spatial transformation, which guarantees topology preservation and invertibility of transformation, but registration accuracy is low. Therefore, a diffeomorphic deformation registration with CNN is proposed, based on a symmetric architecture, simultaneously estimating forward and inverse deformation fields. CNN with Efficient Channel Attention is used to better capture the spatial relationship. Deformation fields are optimized explicitly and implicitly to enhance the invertibility of transformations. An extensive experimental evaluation is performed using two 3D datasets. The proposed method is compared with different state‐of‐the‐art methods. The experimental results show excellent registration accuracy while better guaranteeing the diffeomorphic transformation.

Abstract Unmanned aerial vehicle insulator detection that aims to recognize defective insulators from transmission lines has made significant progress in recent years. However, it still faces challenges, such as the complex background of aerial images and the small memory of unmanned aerial vehicles. This paper proposes a refined insulator detection algorithm that integrates the attention mechanism in YOLOv8 to improve the feature extraction ability. Specifically, this paper introduces a fast vision transformers structure in the you only look once (YOLO) v8 backbone section to enhance feature extraction by capturing local and global features. Additionally, the global attention mechanism is incorporated in the neck for additional feature extraction by merging comprehensive spatial and channel information into the output. Furthermore, we amalgamate depth‐wise convolution, graph convolution, and residual operation in the global attention mechanism module. This design can mitigate the issues of gradient vanishing or exploding and meanwhile enhance the distinction between spatial attention and channel attention. The proposed model is then applied to a public dataset and a set of real images from a specific power station, and the detection results show that it outperforms many competitors in terms of accuracy, efficiency, and memory size.

Abstract Taken in low‐light intensity conditions, image with low brightness affects processing precision. In this article, the Gamma Function based on the brightness average and weighted fusion method according to gray entropy is proposed, which is combined with the improved Retinex algorithm. First, the maximum values of R, G, and B channels in original image are extracted to generate the primary illumination map. Second, the illumination map is optimized and adjusted via the Gamma correction function based on the average brightness value. Finally, the illumination map and detail layer are fused by a weighted fusion algorithm of gray entropy to obtain the reflection map. Reflection maps are used as enhancement. The algorithm proposed in this article can improve the brightness and maintain light distribution in the original image with higher precision and less color distortion.

Abstract When detecting the objects in videos, motion always leads to object deterioration, like blurring and occlusion, as well as the strange state of the object's shape and posture. Consequently, the detection of video frames will lead to a decline in accuracy by using the image object detection model. This paper proposes an online video object detection method based on the one‐stage detector YOLOx. First, the module for space–time feature aggregation is given, which uses the space–time information of past frames to enhance the feature quality of the current frame. Then, the module for result reuse is given, which incorporates the detection results of past frames to improve the detection stability of the current frame. By these two modules, the trade‐off between accuracy and speed of video object detection could be achieved. Experimental results on the ImageNet VID show the improvement of speed and accuracy of the proposed method.

Abstract Cross‐dataset object detection methods can adapt to the needs of rapid category expansion in object detection tasks. However, these methods are prone to generate dataset‐aware errors with false alarm objects. This study is aimed to address these issues. A box‐separated multiple‐head module and box‐separated loss function based on the YOLOv8 network are devised to achieve cross‐dataset object detection. Additionally, a sameclass‐aware fusion module to avoid gradient conflicts due to cross‐category conflicts is developed. A multiple‐head fusion module is devised to reduce the number of false alarm objects caused by dataset‐aware errors. A global class‐aware sampler is also designed to adapt to the impact of the imbalanced number of categories and training samples across datasets. The effectiveness of the box‐separated multiple‐head module is verified using cross‐datasets built using the COCO, WiderFace, WiderPerson, and OpenImages V4 datasets. Extensive experiments demonstrate the efficiency and precision of the proposed method.

Abstract Effective extraction of building from remote sensing images requires both global and local information. Despite convolutional neural networks (CNNs) excelling at capturing local details, their intrinsic focus on local operations poses challenge in effectively extracting global features, especially in the context of large‐scale buildings. In contrast, transformers excel at capturing global information, but compared to CNNs, they tend to overly rely on large‐scale datasets and pre‐trained parameters. To tackle the challenge, this paper presents the multi‐scale and dual‐branch feature fusion network (MDTrans). Specifically, the CNN and transformer branches are integrated in a dual‐branch parallel manner during both encoding and decoding stages, local information for small‐scale buildings is extracted by utilizing Dense Connection Blocks in the CNN branch, while crucial global information for large‐scale buildings is effectively captured through Swin Transformer Block in the transformer branch. Additionally, Dual Branch Information Fusion Block is designed to fuse local and global features from the two branches. Furthermore, Multi‐Convolutional Block is designed to further enhance the feature extraction capability of buildings with different sizes. Through extensive experiments on the WHU, Massachusetts, and Inria building datasets, MDTrans achieves intersection over union (IoU) scores of 91.36%, 64.69%, and 79.25%, respectively, outperforming other state‐of‐the‐art models.

Abstract Considering the challenges of low multi‐object detection accuracy and difficulty in identifying small targets caused by challenging environmental conditions including irregular lighting patterns and ambient noise levels in the mining environment with autonomous electric locomotives. A new network model based on SOD−YOLOv5s−4L has been proposed to detect multi‐objects for autonomous electric locomotives in underground coal mines. Improvements have been applied in YOLOv5s to construct the SOD−YOLOv5s−4L model, by introducing the SIoU loss function to address the mismatch between real and predicted bounding box directions, facilitating the model to learn target position information more efficiently. This research introduces a decoupled head to enhance feature fusion and improve the positioning precision of the network model, enabling rapid capture of multi‐scale target features. Furthermore, the detection capability of the model has been increased by introducing the small target detection layer which is developed by increasing the number of detection layers from three to four. The experimental results on multiple object detection dataset show that the proposed model achieves significant improvement in mean average precision (mAP) of almost 98% for various types of targets and an average precision (AP) of nearly 99% for small targets on the other hand it achieves 5.19% (mAP) and 9.79% (AP) compared to the YOLOv5s model. Furthermore, comparative analysis with other models like YOLOv7 and YOLOv8 shows that the proposed model has superior performance in terms of object detection.