Down-sampling Operations Research Articles

RPDNet: A reconstruction-regularized parallel decoders network for rectal tumor and rectum co-segmentation

Accurate segmentation of rectal cancer tumor and rectum in magnetic resonance imaging (MRI) is significant for tumor precise diagnosis and treatment plans determination. Variable shapes and unclear boundaries of rectal tumors make this task particularly challenging. Only a few studies have explored deep learning networks in rectal tumor segmentation, which mainly adopt the classical encoder-decoder structure. The frequent downsampling operations during feature extraction result in the loss of detailed information, limiting the network's ability to precisely capture the shape and boundary of rectal tumors. This paper proposes a Reconstruction-regularized Parallel Decoder network (RPDNet) to address the problem of information loss and obtain accurate co-segmentation results of both rectal tumor and rectum. RPDNet initially establishes a shared encoder and parallel decoders framework to fully utilize the common knowledge between two segmentation labels while reducing the number of network parameters. An auxiliary reconstruction branch is subsequently introduced by calculating the consistency loss between the reconstructed and input images to preserve sufficient anatomical structure information. Moreover, a non-parameter target-adaptive attention module is proposed to distinguish the unclear boundary by enhancing the feature-level contrast between rectal tumors and normal tissues. The experimental results indicate that the proposed method outperforms state-of-the-art approaches in rectal tumor and rectum segmentation tasks, with Dice coefficients of 84.91 % and 90.36 %, respectively, demonstrating its potential application value in clinical practice.

ThyFusion: A lightweight attribute enhancement module for thyroid nodule diagnosis using gradient and frequency-domain awareness

Accurate classification of thyroid nodules is a critical step in clinical diagnosis and plays a crucial role in guiding subsequent treatment planning. However, current deep learning methods lack the ability to effectively perceive the attributes of thyroid nodules. Specifically, convolutional neural network-based methods struggle to capture fine-grained attributes, such as echogenic foci, due to repeated downsampling operations. On the other hand, Transformer-based methods partition the image into blocks, which hinders the capture of continuous attributes such as margins, and require a large number of parameters. To overcome these limitations, this paper proposes a novel lightweight attribute enhancement module called ThyFusion. Firstly, a multi-scale Gaussian filtering attention module is developed to accurately capture fine-grained information. Secondly, a frequency-domain global regulation module is proposed to regulate global features and capture coarse-grained attributes. Finally, a cross domain mutual learning module is presented to fuse fine-grained and coarse-grained attributes. Extensive experiments were conducted to demonstrate the effectiveness of ThyFusion, and it was compared against 15 benchmark methods. ThyFusion achieves higher accuracy, F1 score, recall, and precision (87.18%, 87.12%, 87.36% and 87.10%). ThyFusion is a plug-and-play lightweight module that can be embedded behind any layer of a backbone network to enhance the accuracy of thyroid nodule classification. The advantages of ThyFusion in thyroid nodule classification can greatly streamline the process of automated diagnosis and improve the accuracy of thyroid diagnosis.

FBSM: Foveabox-based boundary-aware segmentation method for green apples in natural orchards

Image segmentation of target fruits is an essential part of machine vision systems, aiming to facilitate more accurate early fruit measurement and machine harvesting in natural orchard environments. Given that instance-level pooling and down-sampling operations in conventional segmentation models often lead to the loss of detailed information, resulting in coarse partitioning masks, we endeavor to restore the boundary information of the masks. To achieve high-quality fruit segmentation with clear boundaries and refined masks, a Foveabox-based boundary-aware segmentation model (FBSM) is constructed by adding a multi-stage mask prediction head incorporating fine-grained features to the anchor-free detection model FoveaBox to realize accurate segmentation of green apple fruits. At each stage, a bi-layer fusion structure (BFS) is employed to fully fuse the fine-grained features in a double-fusion manner to guide the subsequent instance mask prediction. Finally, the boundary recovery module (BRM) is leveraged to recover the lost boundary detail information and obtain more accurate boundaries for the continuous optimization process of the fruit instance mask. Experimental results demonstrate that the proposed FBSM model achieves a mean average precision (mAP) of 60.7 % for green apple instance segmentation in unstructured natural orchard environments, surpassing traditional instance segmentation models. Moreover, it enhances base model detection accuracy by 1.3 % while striking a better balance between detection and segmentation performance.

Advancing image inpainting efficiency: An exploration of pixel and channel split operations

Deep learning-based image inpainting techniques have achieved unprecedented results using encoder–decoder structures to recover complex missing areas of an image. Recent inpainting models use additional information or networks (e.g., landmarks, edges, styles, and filters) to realize improved restoration performance, but at the cost of increased computational resources. To improve the relationship between inpainting performance and the number of model parameters, researchers have investigated efficient structural approaches such as recurrent and residual connection structures. However, these methods are difficult to apply in the general encoder–decoder structure. In this study, we explored the downsampling and upsampling operations associated with an encoder–decoder structure. We propose two novel split operations: the pixel-split operation (PSO) and channel-split operation (CSO). The proposed PSO transfers image features from high to low resolution with two dilation rate effects and a similar number of parameters as existing downsampling operations. Conversely, the proposed CSO increases the image resolution using only one-fourth the number of parameters of existing upsampling operations. The restoration performance and efficiency of the proposed model were evaluated in terms of five metrics on public datasets, e.g., the Places2 and CelebA datasets, to validate our proposed operations’ contribution to inpainting performance. We achieved state-of-the-art performance and reduced the size of the parameters by 20%. An ablation study was conducted to confirm the effect of each operation on the CelebA-HQ dataset. Results indicated that these split operations exhibit an advanced relationship between inpainting performance and optimization of the model parameters. The corresponding codes are available online (https://github.com/MrCAIcode/Split_operation_for_inpainting).

Boundary-Aware Gradient Operator Network for Medical Image Segmentation.

Medical image segmentation is a crucial task in computer-aided diagnosis. Although convolutional neural networks (CNNs) have made significant progress in the field of medical image segmentation, the convolution kernels of CNNs are optimized from random initialization without explicitly encoding gradient information, leading to a lack of specificity for certain features, such as blurred boundary features. Furthermore, the frequently applied down-sampling operation also loses the fine structural features in shallow layers. Therefore, we propose a boundary-aware gradient operator network (BG-Net) for medical image segmentation, in which the gradient convolution (GConv) and the boundary-aware mechanism (BAM) modules are developed to simulate image boundary features and the remote dependencies between channels. The GConv module transforms the gradient operator into a convolutional operation that can extract gradient features; it attempts to extract more features such as images boundaries and textures, thereby fully utilizing limited input to capture more features representing boundaries. In addition, the BAM can increase the amount of global contextual information while suppressing invalid information by focusing on feature dependencies and the weight ratios between channels. Thus, the boundary perception ability of BG-Net is improved. Finally, we use a multi-modal fusion mechanism to effectively fuse lightweight gradient convolution and U-shaped branch features into a multilevel feature, enabling global dependencies and low-level spatial details to be effectively captured in a shallower manner. We conduct extensive experiments on eight datasets that broadly cover medical images to evaluate the effectiveness of the proposed BG-Net. The experimental results demonstrate that BG-Net outperforms the state-of-the-art methods, particularly those focused on boundary segmentation.

BMCS-Net: A Bi-directional multi-scale cascaded segmentation network based on transformer-guided feature Aggregation for medical images

convolutional neural networks (CNNs) show great potential in medical image segmentation tasks, and can provide reliable basis for disease diagnosis and clinical research. However, CNNs exhibit general limitations on modeling explicit long-range relation, and existing cures, resorting to building deep encoders along with aggressive downsampling operations, leads to loss of localized details. Transformer has naturally excellent ability to model the global features and long-range correlations of the input information, which is strongly complementary to the inductive bias of CNNs. In this paper, a novel Bi-directional Multi-scale Cascaded Segmentation Network, BMCS-Net, is proposed to improve the performance of medical segmentation tasks by aggregating these features obtained from Transformers and CNNs branches. Specifically, a novel feature integration technique, termed as Two-stream Cascaded Feature Aggregation (TCFA) module, is designed to fuse features in two-stream branches, and solve the problem of gradual dilution of global information in the network. Besides, a Multi-Scale Expansion-Aware (MSEA) module based on the convolution of feature perception and expansion is introduced to capture context information, and further compensate for the loss of details. Extensive experiments demonstrated that BMCS-Net has an excellent performance on both skin and Polyp segmentation datasets.

Efficient Mixed-Type Wafer Defect Pattern Recognition Based on Light-Weight Neural Network.

Wafer defect pattern recognition can help engineers improve the production process of semiconductor chips. In real industrial scenarios, the recognition of mixed-type wafer defects is difficult and the production scale of semiconductor wafers is large, which requires high accuracy and speed in wafer defect pattern recognition. This study proposes a light-weight neural network model to efficiently recognize mixed-type wafer defects. The proposed model is constructed via inverted residual convolution blocks with attention mechanisms and large kernel convolution downsampling layers. The inference speed of the inverted residual convolution block is fast, and the attention mechanism can enhance feature extraction capabilities. Large kernel convolutions help the network retain more important feature information during downsampling operations. The experimental results on the real Mixed-type WM38 dataset show that the proposed model achieves a recognition accuracy of 98.69% with only 1.01 M parameters. Compared with some popular high-performance models and light-weight models, our model has advantages in both recognition accuracy and inference speed. Finally, we deploy the model as a TensorRT engine, which significantly improves the inference speed of the model, enabling it to process more than 1300 wafer maps per second.

DCNet: A lightweight retinal vessel segmentation network

Retinal vessel segmentation is a crucial focus within the realm of medical image analysis, playing a pivotal role in early disease diagnosis, notably retinopathy. Deep learning has exhibited remarkable segmentation capabilities for retinal blood vessels, leveraging the advantages of contextual feature learning. However, there are still some shortcomings in fine retinal vessel segmentation due to the loss of semantic information due to too many pooling operations or limited receptive fields due to fewer pooling operations. In response to the nuanced balance required for expanding the receptive field while preserving information integrity during multiple downsampling operations, this paper introduces DCNet (Dilated Convolution Net), a novel lightweight three-layer dilated-convolution-based network tailored for retinal blood vessel segmentation. This three-layer architecture autonomously extracts crucial segmentation features from various levels of the feature map. Each layer comprises a dilated convolution Positive Sequence Block (PSB) and a dilated convolution Reverse Sequence Block (RSB). The dilated convolution operation is strategically exploited for its capacity to extend the receptive field, facilitating effective feature information extraction. Simultaneously, to alleviate semantic information loss within the deep network's feature map, this paper proposes the Nonlinear Feature Extraction Module (NFEM) to supplement shallow network feature information. Furthermore, to comprehensively leverage information from various scale features, a Feature Fusion Module (FFM) is introduced for multiscale vascular feature extraction, ultimately enhancing segmentation accuracy. DCNet undergoes rigorous evaluation on four publicly accessible retinal vascular datasets – DRIVE, STARE, CHASE_DB1, and HRF. Experimental results unequivocally demonstrate that DCNet achieves superior segmentation performance with fewer model parameters compared to existing state-of-the-art methods. The code for DCNet can be accessed on the following website: https://github.com/ChunhuiYu1/DCNet.

CSAN-UNet: Channel Spatial Attention Nested UNet for Infrared Small Target Detection

Segmenting small infrared targets presents a significant challenge for traditional image processing architectures due to the inherent lack of texture, minimal shape information, and their sparse pixel representation within images. The conventional UNet architecture, while proficient in general segmentation tasks, inadequately addresses the nuances of small infrared target segmentation due to its reliance on downsampling operations, such as pooling, which often results in the loss of critical target information. This paper introduces the Channel Spatial Attention Nested UNet (CSAN-UNet), an innovative architecture designed specifically to enhance the detection and segmentation of small infrared targets. Central to CSAN-UNet’s design is the Cascaded Channel and Spatial Convolutional Attention Module (CSCAM), a novel component that adaptively enhances multi-level features and mitigates the loss of target information attributable to downsampling processes. Additionally, the Channel-priority and Spatial Attention Cascade Module (CPSAM) represents another pivotal advancement within CSAN-UNet, prioritizing channel-level adjustments alongside spatial attention mechanisms to efficiently extract deep semantic information pertinent to small infrared targets. Empirical validation conducted on two public datasets confirms that CSAN-UNet surpasses existing state-of-the-art algorithms in segmentation performance, while simultaneously reducing computational overhead.

MAFormer: A transformer network with multi-scale attention fusion for visual recognition

Vision Transformer and its variants have demonstrated great potential in various computer vision tasks. However conventional vision transformers often focus on global dependency at a coarse level, which results in a learning challenge on global relationships and fine-grained representation at a token level. In this paper, we introduce Multi-scale Attention Fusion into transformer (MAFormer), which explores local aggregation and global feature extraction in a dual-stream framework for visual recognition. We develop a simple but effective module to explore the full potential of transformers for visual representation by learning fine-grained and coarse-grained features at a token level and dynamically fusing them. Our Multi-scale Attention Fusion (MAF) block consists of: i) a local window attention branch that learns short-range interactions within windows, aggregating fine-grained local features; ii) global feature extraction through a novel Global Learning with Down-sampling (GLD) operation to efficiently capture long-range context information within the whole image; iii) a fusion module that self-explores the integration of both features via attention. Our MAFormer achieves state-of-the-art results on several common vision tasks. In particular, MAFormer-L achieves 85.9% Top-1 accuracy on ImageNet, surpassing CSWin-B and LV-ViT-L by 1.7% and 0.6% respectively. On MSCOCO, MAFormer outperforms the prior art CSWin by 1.7% mAPs on object detection and 1.4% on instance segmentation with similar-sized parameters. With the performance, MAFormer demonstrates the ability to generalize across various visual benchmarks and prospects as a general backbone for different self-supervised pre-training tasks in the future.

The Squish Map and the $SL_2$ Double Dimer Model

A plane partition, whose 3D Young diagram is made of unit cubes, can be approximated by a "coarser" plane partition, made of cubes of side length 2. Indeed, there are two such approximations obtained by "rounding up" or "rounding down" to the nearest cube. We relate this coarsening (or downsampling) operation to the squish map introduced by the second author in earlier work. We exhibit a related measure-preserving map between the dimer model on the honeycomb graph, and the SL2 double dimer model on a coarser honeycomb graph; we compute the most interesting special case of this map, related to plane partition q-enumeration with 2-periodic weights. As an application, we specialize the weights to be certain roots of unity, obtain novel generating functions (some known, some new, and some conjectural) that (-1)-enumerate certain classes of pairs of plane partitions according to how their dimer configurations interact.

MpMsCFMA-Net: Multi-path Multi-scale Context Feature Mixup and Aggregation Network for medical image segmentation

Automatic and accurate medical image segmentation is a crucial step for clinical diagnosis and treatment planning of diseases. The advanced convolutional neural network (CNN) approaches based on the encoder–decoder structure have achieved state-of-the-art performances in many different medical image segmentation tasks. However, existing networks have insufficient capability of extracting the context information in each encoding stage, so they cannot effectively perceive multi-scale objects in images. In addition, the continuous down-sampling and convolution operations in the encoding stage lead to much loss of the detailed information, resulting in poor segmentation performance. In this paper, we propose a Multi-path Multi-scale Context Feature MixUp and Aggregation Network (named MpMsCFMA-Net) which fuses and aggregates multi-path features with multi-scale context information to address these issues. Based on the encoder–decoder structure, we first design the encoder to encode the semantic and detailed information of input images and introduce multi-scale context extraction module in each encoding stage. Furthermore, we design multiple features mixup module between the encoder and the decoder, aiming at providing different levels of global context information for the decoder by reconstructing skip-connection. Finally, we introduce the decoder with deeper features aggregation to better fuse multi-scale context information across layers. Experimental results on four public medical image datasets confirm that our proposed network achieves promising results and outperforms other state-of-the-art methods in most of evaluation metrics. The source code will be publicly available at https://github.com/tricksterANDthug/MpMsCFMA-Net.

Information-guided signal multi-granularity contrastive feature learning for fault diagnosis with few labeled data

Fault diagnosis is important for isolating the root causes and supporting preventive maintenance of an equipment. However, the lack of labeled fault samples often limits the performance of fault diagnosis in practical situations. In this paper, we propose an information-guided signal multi-granularity contrastive feature learning (IMCFL) method with few labeled data to identify fault types. Unlike manual selection of data augmentation in traditional contrastive learning, IMCFL adaptively generates task-relevant augmented views for contrastive learning by differentiable automatic augmentation guided by information from few labeled (no more than 5%) data. Furthermore, a multi-granularity contrastive loss is constructed for the feature extractor to generate discriminative feature representation considering the characteristics of the signals. By exploring instance-level relationships and temporal relationships within the signal in multiple successive down-sampling operations, features of different granularities can be adequately extracted. After completing the contrastive learning, fault diagnosis is accomplished by a simple classifier without fine-tuning. Experiments on bearing and gear datasets demonstrate the diagnostic accuracy of the proposed method compared to other methods. Further transfer learning experiments show that the proposed method has the ability to be applied to different devices for fault diagnosis through knowledge transfer of its pre-trained powerful feature extractors.

Wavelet-based Auto-Encoder for simultaneous haze and rain removal from images

Noise introduced due to weather can reduce the efficiency of computer vision applications as the visibility of the objects in images is greatly affected. Haze and rain are the most common weather conditions seen in nature. However, most of the algorithms found in the literature apply rain and haze removal approaches separately. To this end, in this paper, we propose a novel Wavelet-based deep Auto-encoder, called WAE, for simultaneously removing the haze and rain effects from images. The proposed network uses wavelet transformation and inverse wavelet transformation as an alternative to down-sampling and up-sampling operations, respectively, in order to add sparsity to the network. By training the model on both spatial and frequency domains, it learns non-stationary features that are found to be useful to remove haze and rain effects from images. The proposed model is tested on several rain and haze-affected image datasets, and it performs well in terms of standard evaluation metrics like structural similarity index measure and peak signal-to-noise ratio. The code can be found at : https://github.com/asfakali/WAE.git.

Detection algorithm for dense small objects in high altitude image

The high altitude images captured by drones often contain small-sized objects, which pose several challenges including significant scale variations and dense clustering. To tackle these challenges in small object detection, this paper presents EM-YOLO (Efficient Switching Network and Multi-Scale Feature Fusion Reinforced YOLO for Small Object Detection), an enhanced algorithm based on the YOLOv5s architecture. Firstly, to enhance the network's adaptability to varying object feature scales, the EM-YOLO redesigns the feature extraction backbone network by using different dilation rates in convolutions within the backbone. This flexible structure allows for improved extraction of object features, enhancing the network's feature extraction capability. Secondly, additional detection layer for small objects and combines it with the dynamic detection head. This combination addresses the issue of small object information loss due to multiple down-sampling operations on the feature maps. Finally, Focal-EIoU Loss is employed to enhance regression accuracy in EM-YOLO. Extensive experiments were conducted on the VisDrone2019-DET dataset. The results of these experiments demonstrate that the improved algorithm achieved a significant increase in performance. Specifically, the algorithm achieves 43.5% and 25.1% on the mAP0.5 and mAP0.5:0.95 metrics, which are 13.9% and 8.9% higher than YOLOv5s algorithm, respectively. Furthermore, comparative experiments with other mainstream algorithms indicate that EM-YOLO outperforms them in the task of small object detection.

DPNet: Dual-Path Network for Real-Time Object Detection With Lightweight Attention.

The recent advances in compressing high-accuracy convolutional neural networks (CNNs) have witnessed remarkable progress in real-time object detection. To accelerate detection speed, lightweight detectors always have few convolution layers using a single-path backbone. Single-path architecture, however, involves continuous pooling and downsampling operations, always resulting in coarse and inaccurate feature maps that are disadvantageous to locate objects. On the other hand, due to limited network capacity, recent lightweight networks are often weak in representing large-scale visual data. To address these problems, we present a dual-path network, named DPNet, with a lightweight attention scheme for real-time object detection. The dual-path architecture enables us to extract in parallel high-level semantic features and low-level object details. Although DPNet has a nearly duplicated shape with respect to single-path detectors, the computational costs and model size are not significantly increased. To enhance representation capability, a lightweight self-correlation module (LSCM) is designed to capture global interactions, with only a few computational overheads and network parameters. In the neck, LSCM is extended into a lightweight cross correlation module (LCCM), capturing mutual dependencies among neighboring scale features. We have conducted exhaustive experiments on MS COCO, Pascal VOC 2007, and ImageNet datasets. The experimental results demonstrate that DPNet achieves a state-of-the-art trade off between detection accuracy and implementation efficiency. More specifically, DPNet achieves 31.3% AP on MS COCO test-dev, 82.7% mAP on Pascal VOC 2007 test set, and 41.6% mAP on ImageNet validation set, together with nearly 2.5M model size, 1.04 GFLOPs, and 164 and 196 frames/s (FPS) FPS for [Formula: see text] input images of three datasets.

Robust Image Steganography Against General Downsampling Operations With Lossless Secret Recovery

Resisting the operations in lossy channels is a challenge for image steganography. In this paper, we propose a novel robust steganographic method to resist the image downsampling operation. Unlike the existing schemes, our method guarantees lossless secret recovery from the stego-image after general image downsampling operations, which considers the undetectability of the stego-images on both sides (sender and receiver) of the lossy channel. We first downsample the cover image to get its downsampled version on the receiver side, and select a set of embeddable pixels (i.e., the pixels that can be modified for data embedding) from the downsampled image. Then, we generate a stego-image on the receiver side (termed as the receiver stego-image) such that the distortion caused by the data embedding is minimized. Based on the receiver stego-image, we modify the cover image to produce the stego-image on the sender side (termed as the sender stego-image). The modification takes the embedding cost into account and makes sure that the downsampled version of the sender stego-image produces the same embeddable pixels as the receiver stego-image. Experimental results show that our method performs significantly better than the existing schemes in terms of robustness and undetectability for resisting general image downsampling operations.

TNPC: Transformer-based network for point cloud classification

Point cloud classification has emerged as a vital research area in several emerging applications, including robotics and autonomous driving. However, discriminative feature learning has long been a challenging issue owing to the irregularity and disorder of point clouds. Recently, although Transformer-based methods can achieve high accuracy in point cloud learning, plenty of Transformer layers bring huge computation and memory consumption. This paper presents a novel hierarchical local–global framework based on Transformer network for point clouds, named TNPC. TNPC contains two serial Stages implementing downsampling operation, and each Stage is composed of a local feature extracting (LFE) block and a global feature extracting (GFE) block, which can reduce computation and memory consumption obviously. LFE block consists of two parallel branches, namely Transformer branch and shared multilayer perceptron (SMLP) branch, which are designed to learn relevant feature of any two points and local high-dimensional semantic features of each point between sampling centroid and its neighborhoods, respectively. The proposed two parallel branches not only improve the feature extraction effect, but also reduce the computation and memory consumption. GFE block consists of a center point contact (CPC) module and a global point cloud transformer layer (PCTL) module, which can improve the effect of global features extracting without adding the number of parameters and computation. The performance of our method is validated experimentally on ModelNet40 and ScanObjectNN datasets. Our method improves the mean accuracy on each category (mAcc) to 91.6% and 79.8% on the ModelNet40 dataset and ScanObjectNN dataset, respectively. In terms of efficiency, our method leads to a significant reduction, with only 4.73MB parameters and only 1.91GB floating-point operations (FLOPs). Experimental results demonstrate that the proposed method achieves state-of-the-art performance on classification accuracy and efficiency.

A three-layer decomposition method based on structural texture perception for fusion of CT and MRI images

The fusion of computed tomography (CT) images and magnetic resonance imaging (MRI) images has made essential contribution to clinical diagnosis because of its ability of complementary visual information enhancement and redundancy elimination. However, some methods that involve upsampling and downsampling operations may lose information during image processing, which affects the fusion results. In this paper, a medical image fusion algorithm based on structural texture perception is proposed. The proposed method demonstrates enhanced performance in preserving both detailed texture information and energy information of source images. First, the source image is decomposed into a signal strength layer, a base layer, and a texture layer based on the proposed three-layer decomposition framework. Then, the signal strength layers are fused using a saliency detection method based on iterative least squares. The texture layer is fused using the principle of spatial frequency maximization. For the fusion of the base layer, a non-linear function is designed to calculate the fusion weights. Finally, the final fusion result is obtained through image reconstruction. The proposed algorithm is compared with nine state-of-the-art fusion algorithms to verify its superiority. Experimental results show that the proposed method can effectively preserve the intensity information of CT images and the detailed texture of MRI images.

Remote Sensing Target Detection Algorithm based on CBAM-YOLOv5

With the continuous improvement of traditional target detection algorithms, remote sensing targets have become a research hotspot. Aiming at the problems of low recognition accuracy caused by small target size and high background complexity in remote sensing images taken from overhead view, this paper proposes a remote sensing aircraft detection algorithm based on CBAM-YOLOv5. By introducing the lightweight convolutional attention module CBAM module into the YOLOv5 network, the feature extraction capability of the algorithm is improved to solve the problem that the small-size remote sensing target has little or even lost information on the feature map after multiple downsampling operations. The mAP of the proposed algorithm reaches 93.1%, which is 1% higher than that of the original algorithm, and the recognition of small-size remote sensing targets has been significantly improved.