Deformable Convolution Research Articles

Deep learning architecture with shunted transformer and 3D deformable convolution for voxel-level dose prediction of head and neck tumors.

Intensity-modulated radiation therapy (IMRT) has been widely used in treating head and neck tumors. However, due to the complex anatomical structures in the head and neck region, it is challenging for the plan optimizer to rapidly generate clinically acceptable IMRT treatment plans. A novel deep learning multi-scale Transformer (MST) model was developed in the current study aiming to accelerate the IMRT planning for head and neck tumors while generating more precise prediction of the voxel-level dose distribution. The proposed end-to-end MST model employs the shunted Transformer to capture multi-scale features and learn a global dependency, and utilizes 3D deformable convolution bottleneck blocks to extract shape-aware feature and compensate the loss of spatial information in the patch merging layers. Moreover, data augmentation and self-knowledge distillation are used to further improve the prediction performance of the model. The MST model was trained and evaluated on the OpenKBP Challenge dataset. Its prediction accuracy was compared with three previous dose prediction models: C3D, TrDosePred, and TSNet. The predicted dose distributions of our proposed MST model in the tumor region are closest to the original clinical dose distribution. The MST model achieves the dose score of 2.23 Gy and the DVH score of 1.34 Gy on the test dataset, outperforming the other three models by 8%-17%. For clinical-related DVH dosimetric metrics, the prediction accuracy in terms of mean absolute error (MAE) is 2.04% for , 1.54% for , 1.87% for , 1.87% for , 1.89% for , respectively, superior to the other three models. The quantitative results demonstrated that the proposed MST model achieved more accurate voxel-level dose prediction than the previous models for head and neck tumors. The MST model has a great potential to be applied to other disease sites to further improve the quality and efficiency of radiotherapy planning.

Non-Contact Blood Pressure Estimation From Radar Signals by a Stacked Deformable Convolution Network.

This study introduces a contactless blood pressure monitoring approach that combines conventional radar signal processing with novel deep learning architectures. During the preprocessing phase, datasets suitable for synchronization are created by integrating Kalman filtering, multiscale bandpass filters, and a periodic extraction method in the time domain. These data comprise data on chest micro variations, encapsulating a complex array of physiological and biomedical information reflective of cardiac micromotions. The Radar-based Stacked Deformable convolution Network (RSD-Net) integrates channel and spatial self attention mechanisms within a deformable convolutional framework to enhance feature extraction from radar signals. The network architecture systematically employs deformable convolutions for initial deep feature extraction from individual signals. Subsequently, continuous blood pressure estimation is conducted using self attention mechanisms on feature map from single source coupled with multi-feature map channel attention. The performance of model is corroborated via the open-source dataset procured using a non-invasive 24 GHz six-port continuous wave radar system. The dataset, encompassing readings from 30 healthy individuals subjected to diverse conditions including rest, the Valsalva maneuver, apnea, and tilt-table examinations. It serves to substantiate the validity and resilience of the proposed method in the non-contact assessment of continuous blood pressure. Evaluation metrics reveal Pearson correlation coefficients of 0.838 for systolic and 0.797 for diastolic blood pressure predictions. The Mean Error (ME) and Standard Deviation (SD) for systolic and diastolic blood pressure measurements are -0.32 ±6.14 mmHg and -0.20 ±5.50 mmHg, respectively. The ablation study assesses the contribution of different structural components of the RSD-Net, validating their significance in the overall of model performance.

Intelligent skin lesion segmentation using deformable attention Transformer U-Net with bidirectional attention mechanism in skin cancer images.

In recent years, the increasing prevalence of skin cancers, particularly malignant melanoma, has become a major concern for public health. The development of accurate automated segmentation techniques for skin lesions holds immense potential in alleviating the burden on medical professionals. It is of substantial clinical importance for the early identification and intervention of skin cancer. Nevertheless, the irregular shape, uneven color, and noise interference of the skin lesions have presented significant challenges to the precise segmentation. Therefore, it is crucial to develop a high-precision and intelligent skin lesion segmentation framework for clinical treatment. A precision-driven segmentation model for skin cancer images is proposed based on the Transformer U-Net, called BiADATU-Net, which integrates the deformable attention Transformer and bidirectional attention blocks into the U-Net. The encoder part utilizes deformable attention Transformer with dual attention block, allowing adaptive learning of global and local features. The decoder part incorporates specifically tailored scSE attention modules within skip connection layers to capture image-specific context information for strong feature fusion. Additionally, deformable convolution is aggregated into two different attention blocks to learn irregular lesion features for high-precision prediction. A series of experiments are conducted on four skin cancer image datasets (i.e., ISIC2016, ISIC2017, ISIC2018, and PH2). The findings show that our model exhibits satisfactory segmentation performance, all achieving an accuracy rate of over 96%. Our experiment results validate the proposed BiADATU-Net achieves competitive performance supremacy compared to some state-of-the-art methods. It is potential and valuable in the field of skin lesion segmentation.

A frequency selection network for medical image segmentation

Existing medical image segmentation methods may only consider feature extraction and information processing in spatial domain, or lack the design of interaction between frequency information and spatial information, or ignore the semantic gaps between shallow and deep features, and lead to inaccurate segmentation results. Therefore, in this paper, we propose a novel frequency selection segmentation network (FSSN), which achieves more accurate lesion segmentation by fusing local spatial features and global frequency information, better design of feature interactions, and suppressing low correlation frequency components for mitigating semantic gaps. Firstly, we propose a global-local feature aggregation module (GLAM) to simultaneously capture multi-scale local features in the spatial domain and exploits global frequency information in the frequency domain, and achieves complementary fusion of local details features and global frequency information. Secondly, we propose a feature filter module (FFM) to mitigate semantic gaps when we conduct cross-level features fusion, and makes FSSN discriminatively determine which frequency information should be preserved for accurate lesion segmentation. Finally, in order to make better use of local information, especially the boundary of lesion region, we employ deformable convolution (DC) to extract pertinent features in the local range, and makes our FSSN can focus on relevant image contents better. Extensive experiments on two public benchmark datasets show that compared with representative medical image segmentation methods, our FSSN can obtain more accurate lesion segmentation results in terms of both objective evaluation indicators and subjective visual effects with fewer parameters and lower computational complexity.

Trans-DCN: A High-Efficiency and Adaptive Deep Network for Bridge Cable Surface Defect Segmentation

Cables are vital load-bearing components of cable-stayed bridges. Surface defects can lead to internal corrosion and fracturing, significantly impacting the stability of the bridge structure. The detection of surface defects from bridge cable images faces numerous challenges, including shadow disturbances due to uneven lighting and difficulties in addressing multiscale defect features. To address these challenges, this paper proposes a novel and cost-effective deep learning segmentation network, named Trans-DCN, to detect defects in the surface of the bridge cable. The network leverages an efficient Transformer-based encoder and integrates multiscale features to overcome the limitations associated with local feature inadequacy. The decoder implements an atrous Deformable Convolution (DCN) pyramid and dynamically fuses low-level feature information to perceive the complex distribution of defects. The effectiveness of Trans-DCN is evaluated by comparing it with state-of-the-art segmentation baseline models using a dataset comprising cable bridge defect images. Experimental results demonstrate that our network outperforms the state-of-the-art network SegFormer, achieving a 27.1% reduction in GFLOPs, a 1.2% increase in mean Intersection over Union, and a 1.5% increase in the F1 score. Ablation experiments confirmed the effectiveness of each module within our network, further substantiating the significant validity and advantages of Trans-DCN in the task of bridge cable defect segmentation. The network proposed in this paper provides an effective solution for downstream cable bridge image analysis.

Light-YOLO: A Study of a Lightweight YOLOv8n-Based Method for Underwater Fishing Net Detection

Detecting small dark targets underwater, such as fishing nets, is critical to the operation of underwater robots. Existing techniques often require more computational resources and operate under harsh underwater imaging conditions when handling such tasks. This study aims to develop a model with low computational resource consumption and high efficiency to improve the detection accuracy of fishing nets for safe and efficient underwater operations. The Light-YOLO model proposed in this paper introduces an attention mechanism based on sparse connectivity and deformable convolution optimized for complex underwater lighting and visual conditions. This novel attention mechanism enhances the detection performance by focusing on the key visual features of fishing nets, while the introduced CoTAttention and SEAM modules further improve the model’s recognition accuracy of fishing nets through deeper feature interactions. The results demonstrate that the proposed Light-YOLO model achieves a precision of 89.3%, a recall of 80.7%, and an mAP@0.5 of 86.7%. Compared to other models, our model has the highest precision for its computational size and is the lightest while maintaining similar accuracy, providing an effective solution for fishing net detection and identification.

DeMambaNet: Deformable Convolution and Mamba Integration Network for High-Precision Segmentation of Ambiguously Defined Dental Radicular Boundaries.

The incorporation of automatic segmentation methodologies into dental X-ray images refined the paradigms of clinical diagnostics and therapeutic planning by facilitating meticulous, pixel-level articulation of both dental structures and proximate tissues. This underpins the pillars of early pathological detection and meticulous disease progression monitoring. Nonetheless, conventional segmentation frameworks often encounter significant setbacks attributable to the intrinsic limitations of X-ray imaging, including compromised image fidelity, obscured delineation of structural boundaries, and the intricate anatomical structures of dental constituents such as pulp, enamel, and dentin. To surmount these impediments, we propose the Deformable Convolution and Mamba Integration Network, an innovative 2D dental X-ray image segmentation architecture, which amalgamates a Coalescent Structural Deformable Encoder, a Cognitively-Optimized Semantic Enhance Module, and a Hierarchical Convergence Decoder. Collectively, these components bolster the management of multi-scale global features, fortify the stability of feature representation, and refine the amalgamation of feature vectors. A comparative assessment against 14 baselines underscores its efficacy, registering a 0.95% enhancement in the Dice Coefficient and a diminution of the 95th percentile Hausdorff Distance to 7.494.

One-Year-Old Precocious Chinese Mitten Crab Identification Algorithm Based on Task Alignment.

The cultivation of the Chinese mitten crab (Eriocheir sinensis) is an important component of China's aquaculture industry and also a field of concern worldwide. It focuses on the selection of high-quality, disease-free juvenile crabs. However, the early maturity rate of more than 18.2% and the mortality rate of more than 60% make it difficult to select suitable juveniles for adult culture. The juveniles exhibit subtle distinguishing features, and the methods for differentiating between sexes vary significantly; without training from professional breeders, it is challenging for laypersons to identify and select the appropriate juveniles. Therefore, we propose a task-aligned detection algorithm for identifying one-year-old precocious Chinese mitten crabs, named R-TNET. Initially, the required images were obtained by capturing key frames, and then they were annotated and preprocessed by professionals to build a training dataset. Subsequently, the ResNeXt network was selected as the backbone feature extraction network, with Convolutional Block Attention Modules (CBAMs) and a Deformable Convolution Network (DCN) embedded in its residual blocks to enhance its capability to extract complex features. Adaptive spatial feature fusion (ASFF) was then integrated into the feature fusion network to preserve the detailed features of small targets such as one-year-old precocious Chinese mitten crab juveniles. Finally, based on the detection head proposed by task-aligned one-stage object detection, the parameters of its anchor alignment metric were adjusted to detect, locate, and classify the crab juveniles. The experimental results showed that this method achieves a mean average precision (mAP) of 88.78% and an F1-score of 97.89%. This exceeded the best-performing mainstream object detection algorithm, YOLOv7, by 4.17% in mAP and 1.77% in the F1-score. Ultimately, in practical application scenarios, the algorithm effectively identified one-year-old precocious Chinese mitten crabs, providing technical support for the automated selection of high-quality crab juveniles in the cultivation process, thereby promoting the rapid development of aquaculture and agricultural intelligence in China.

SwinDefNet: A Novel Surface Water Mapping Model in Mountain and Cloudy Regions Based on Sentinel-2 Imagery

Surface water plays a pivotal role in the context of climate change, human activities, and ecosystems, underscoring the significance of precise monitoring and observation of surface water bodies. However, the intricate and diverse nature of surface water distribution poses substantial challenges to accurate mapping. The extraction of water bodies from medium-resolution satellite remote sensing images using CNN methods is constrained by limitations in receptive fields and inadequate context modeling capabilities, resulting in the loss of boundary details of water bodies and suboptimal fusion of multi-scale features. The existing research on this issue is limited, necessitating the exploration of novel deep-learning network combinations to overcome these challenges. This study introduces a novel deep learning network combination, SwinDefNet, which integrates deformable convolution and Swin Transformer for the first time. By enhancing the effective receptive field and integrating global semantic information, the model can effectively capture the diverse features of water bodies at various scales, thereby enhancing the accuracy and completeness of water extraction. The model was evaluated on Sentinel-2 satellite images, achieving an overall accuracy of 97.89%, an F1 score of 92.33%, and, notably, an accuracy of 98.03% in mountainous regions. These findings highlight the promising potential of this combined approach for precise water extraction tasks.

LDConv: Linear deformable convolution for improving convolutional neural networks

Neural networks based on convolutional operations have achieved remarkable results in the field of deep learning, but there are two inherent flaws in standard convolutional operations. On the one hand, the convolution operation is confined to a local window, so it cannot capture information from other locations, and its sampled shapes is fixed. On the other hand, the size of the convolutional kernel is fixed to k × k, which is a fixed square shape, and the number of parameters tends to grow squarely with size. Although Deformable Convolution (Deformable Conv) address the problem of fixed sampling of standard convolutions, the number of parameters also tends to grow in a squared manner, and Deformable Conv do not explore the effect of different initial sample shapes on network performance. In response to the above questions, the Linear Deformable Convolution (LDConv) is explored in this work, which gives the convolution kernel an arbitrary number of parameters and arbitrary sampled shapes to provide richer options for the trade-off between network overhead and performance. In LDConv, a novel coordinate generation algorithm is defined to generate different initial sampled positions for convolutional kernels of arbitrary size. To adapt to changing targets, offsets are introduced to adjust the shape of the samples at each position. LDConv corrects the growth trend of the number of parameters for standard convolution and Deformable Conv to a linear growth. Compared to Deformable Conv, LDConv provides richer choices and can be equivalent to deformable convolution when the number of parameters of LDConv is set to the square of K. Differently, this paper also explores the effect of neural networks by using LDConv with the same size and different initial sampling shapes. LDConv completes the process of efficient feature extraction by irregular convolutional operations and brings more exploration options for convolutional sampled shapes. Object detection experiments on representative datasets COCO2017, VOC 7 + 12, and VisDrone-DET2021 fully demonstrate the advantages of LDConv. LDConv is a plug-and-play convolutional operation that can replace the convolutional operation to improve network performance. The code for the relevant tasks can be found at https://github.com/CV-ZhangXin/LDConv.

Stereo matching and 3D reconstruction with NeRF supervision for accurate weight estimation in free-swimming fish

Non-invasive methods of accurate estimation of fish weight are essential for biomass assessment, precision feeding, and effective management in aquaculture. However, the lack of large-scale stereo matching datasets and costly depth sensing equipment in underwater scenarios present significant challenges to accurately measure the size of free-swimming fish. This study introduces a novel NeRF-supervised stereo matching network for precise depth estimation and 3D reconstruction of the fish body. The state-of-the-art neural rendering methods are employed to generate stereo training data from image sequences. In this study, the residual network with deformable convolutions was proposed to extract contextual texture features, significantly enhancing disparity estimation accuracy at the edge of the fish body. Additionally, multiscale spatial features are fused using a feature pyramid network (FPN). To enhance the adaptability of disparity search, we introduce a lightweight Triplet Attention module to capture cross-dimensional dependencies prior to constructing the correlation volume. Experimental results demonstrate the effectiveness of the proposed depth estimation method, achieving an impressive result of 92.35% in the similarity metric, compared to the ground truth. Moreover, we developed a weight estimation model based on bass perimeter using instance segmentation and 3D reconstruction techniques. Notably, a robust correlation (correlation coefficient of 0.98) between fish perimeter and weight is observed, highlighting the potential application of our methods to enhance the accuracy of weight estimation for free-swimming fish.

Deep Learning-Based Automated Cell Detection-Facilitated Meat Quality Evaluation.

Meat consumption is increasing globally. The safety and quality of meat are considered important issues for human health. During evaluations of meat quality and freshness, microbiological parameters are often analyzed. Counts of indicator cells can provide important references for meat quality. In order to eliminate the error of manual operation and improve detection efficiency, this paper proposed a Convolutional Neural Network (CNN) with a backbone called Detect-Cells-Rapidly-Net (DCRNet), which can identify and count stained cells automatically. The DCRNet replaces the single channel of residual blocks with the aggregated residual blocks to learn more features with fewer parameters. The DCRNet combines the deformable convolution network to fit flexible shapes of stained animal cells. The proposed CNN with DCRNet is self-adaptive to different resolutions of images. The experimental results indicate that the proposed CNN with DCRNet achieves an Average Precision of 81.2% and is better than traditional neural networks for this task. The difference between the results of the proposed method and manual counting is less than 0.5% of the total number of cells. The results indicate that DCRNet is a promising solution for cell detection and can be equipped in future meat quality monitoring systems.

A High-Precision Human Fall Detection Model Based on FasterNet and Deformable Convolution

To address the challenges of low accuracy and suboptimal real-time performance in fall detection, caused by lighting variations, occlusions, and complex human poses, a novel fall detection algorithm, FDT-YOLO, has been developed. This algorithm builds upon an improved YOLOv8 framework, featuring significant modifications for improved performance. The C2f module in the backbone network has been replaced with the FasterNet module. This substitution enhances feature reuse effectively and reduces computational complexity. Additionally, a deformable convolution module has been added to the neck section. This helps to decrease missed and false detections significantly, which are often caused by dramatic changes in fall poses. Furthermore, the triplet attention mechanism has been incorporated during multi-scale fusion. This mechanism effectively suppresses background interference, focusing more on the target area, thereby improving detection accuracy and robustness. Experimental results have demonstrated that improvements in FDT-YOLO lead to notable enhancements. The mean Average Precision (mAP) at an Intersection over Union (IoU) threshold of 0.5 has been increased from 94.9% to 96.2%. The mAP for the range of 0.5 to 0.95 has been raised from 84.2% to 85.9%. Additionally, the parameter count has been reduced to 9.9 million, which not only enhances detection accuracy but also significantly reduces the rate of false detections.

PUDCN: two-dimensional phase unwrapping with a deformable convolutional network.

Two-dimensional phase unwrapping is a fundamental yet vital task in optical imaging and measurement. In this paper, what we believe to be a novel deep learning framework PUDCN is proposed for 2D phase unwrapping. We introduce the deformable convolution technique in the PUDCN and design two deformable convolution-related plugins for dynamic feature extraction. In addition, PUDCN adopts a coarse-to-fine strategy that unwraps the phase in the first stage and then refines the unwrapped phase in the second stage to obtain an accurate result. The experiments show that our PUDCN performs better than the existing state-of-the-art. Furthermore, we apply PUDCN to unwrap the phase of optical fibers in optical interferometry, demonstrating its generalization ability.

ASD-YOLO: An aircraft surface defects detection method using deformable convolution and attention mechanism

Aircraft surface defect (ASD) detection is crucial for ensuring flight safety. Addressing challenges such as large-scale variations, irregular shapes, and sample imbalance in ASD detection, this paper proposes a ASD-YOLO network based on YOLOv5. ASD-YOLO incorporates several enhancements to improve recognition capabilities. Firstly, a new deformable convolutional feature extraction module (DCNC3) is designed to better learn defects of different shapes, which is combined with a global attention mechanism (GAM) to pay more attention to defects region information. Secondly, the feature representation of small defects is bolstered by the contextual enhancement module (CEM). Lastly, to alleviate sample imbalance problem, we introduce an exponential sliding average weight function (EMA-Slide). Experimental results on two datasets show improvements in mean Average Precision (mAP) by 5.7% and 3.4%, respectively, surpassing mainstream algorithms and offering a novel approach to ASD detection.

Real-Time Multimodal 3D Object Detection with Transformers

The accuracy and real-time performance of 3D object detection are key factors limiting its widespread application. While cameras capture detailed color and texture features, they lack depth information compared to LiDAR. Multimodal detection combining both can improve results but incurs significant computational overhead, affecting real-time performance. To address these challenges, this paper presents a real-time multimodal fusion model called Fast Transfusion that combines the benefits of LiDAR and camera sensors and reduces the computational burden of their fusion. Specifically, our Fast Transfusion method uses QConv (Quick Convolution) to replace the convolutional backbones compared to other models. QConv concentrates the convolution operations at the feature map center, where the most information resides, to expedite inference. It also utilizes deformable convolution to better match the actual shapes of detected objects, enhancing accuracy. And the model incorporates EH Decoder (Efficient and Hybrid Decoder) which decouples multiscale fusion into intra-scale interaction and cross-scale fusion, efficiently decoding and integrating features extracted from multimodal data. Furthermore, our proposed semi-dynamic query selection refines the initialization of object queries. On the KITTI 3D object detection dataset, our proposed approach reduced the inference time by 36 ms and improved 3D AP by 1.81% compared to state-of-the-art methods.

A Lightweight and Efficient Multi-Type Defect Detection Method for Transmission Lines Based on DCP-YOLOv8.

Currently, the intelligent defect detection of massive grid transmission line inspection pictures using AI image recognition technology is an efficient and popular method. Usually, there are two technical routes for the construction of defect detection algorithm models: one is to use a lightweight network, which improves the efficiency, but it can generally only target a few types of defects and may reduce the detection accuracy; the other is to use a complex network model, which improves the accuracy, and can identify multiple types of defects at the same time, but it has a large computational volume and low efficiency. To maintain the model's high detection accuracy as well as its lightweight structure, this paper proposes a lightweight and efficient multi type defect detection method for transmission lines based on DCP-YOLOv8. The method employs deformable convolution (C2f_DCNv3) to enhance the defect feature extraction capability, and designs a re-parameterized cross phase feature fusion structure (RCSP) to optimize and fuse high-level semantic features with low level spatial features, thus improving the capability of the model to recognize defects at different scales while significantly reducing the model parameters; additionally, it combines the dynamic detection head and deformable convolutional v3's detection head (DCNv3-Dyhead) to enhance the feature expression capability and the utilization of contextual information to further improve the detection accuracy. Experimental results show that on a dataset containing 20 real transmission line defects, the method increases the average accuracy (mAP@0.5) to 72.2%, an increase of 4.3%, compared with the lightest baseline YOLOv8n model; the number of model parameters is only 2.8 M, a reduction of 9.15%, and the number of processed frames per second (FPS) reaches 103, which meets the real time detection demand. In the scenario of multi type defect detection, it effectively balances detection accuracy and performance with quantitative generalizability.

DeformableFishNet: a high-precision lightweight target detector for underwater fish identification

The application of computer vision in fish identification facilitates researchers and managers to better comprehend and safeguard the aquatic ecological environment. Numerous researchers have harnessed deep learning methodologies for studying fish species identification. Nonetheless, this endeavor still encounters challenges such as high computational costs, a substantial number of parameters, and limited practicality. To address these issues, we propose a lightweight network architecture incorporating deformable convolutions, termed DeformableFishNet. Within DeformableFishNet, an efficient global coordinate attention module (EGCA) is introduced alongside a deformable convolution network (EDCN/EC2f), which is grounded in EGCA, to tackle the deformation of fish bodies induced by swimming motions. Additionally, an EC2f-based feature pyramid network (EDBFPN) and an efficient multi-scale decoupling head (EMSD Head) are proposed to extract multi-scale fish features within a lightweight framework. DeformableFishNet was deployed on our freshwater fish dataset, with experimental outcomes illustrating its efficacy, achieving a mean average precision (mAP) of 96.3%. The model comprises 1.7 million parameters and entails 4.7 billion floating-point operations (FLOPs). Furthermore, we validated DeformableFishNet on three public underwater datasets, yielding respective mAPs of 98%, 99.4%, and 83.6%. The experiments show that DeformableFishNet is suitable for underwater identification of various scenes.

Feature aggregation network for small object detection

Due to the miniature scale and limited identifiable features, small objects pose a significant challenge in detection. Improving the accuracy of small object detection is a momentous issue of concern among researchers. Feature pyramid network employs a divide-and-conquer strategy for detecting small objects in low-level networks. However, the limited semantic information in these networks results in suboptimal performance in small object detection. To address this issue, we fully utilize information from all feature levels and propose a Feature Aggregation Network (FAN). We investigate information propagation pathways in neural networks, analyze early fusion and late fusion of features, and introduce a dual top-down pathway that utilizes high-level semantic information to consistently reinforce low-level spatial information. We design a Feature-Aware Module that narrows the semantic gap and steers the network toward learning features that favor small object detection. We employ deformable convolution to accurately locate the boundaries of objects with varying shapes and sizes. FAN can function as a plug-and-play component with minimal computational overhead and be trained end-to-end alongside backbone networks. Extensive experiments are conducted on the COCOs, TinyPerson, and VisDrone datasets. The highly competitive results demonstrate that our approach exhibits robust generalization capabilities and can further improve the accuracy of small object detection.

An accurate measurement of high-reflective objects by using 3D structured light

Using three-dimensional structured light to measure high-reflective objects accurately is a big challenge. Some overexposed and blurred areas in the structured light fringe patterns lead to the loss of semantic and texture information on the object surface, further leading to 3D model reconstruction errors. To solve this problem, a lightweight novel network, Deformable Convolutional and Multi-scale Convolutional Network (DcMcNet), is proposed for repairing highly reflectively distorted regions in sinusoidal fringe patterns and the Gray code binary fringe patterns. DcMcNet uses deformable convolution and multi-scale convolution to realize the effective utilization of multiscale global features in repairing highly reflective distorted regions and can adapt well to the different shapes of these regions. In addition, virtual software is used to acquire large-scale and high-fidelity datasets and a diverse loss function is constructed to better optimize DcMcNet. Depth reconstruction experiments of sinusoidal fringe patterns and point cloud reconstruction experiments of Gray code binary fringe patterns show that our method can repair highly reflective distortion regions with high accuracy, and for aeronautical blades, the MAE of the depth map is reduced from 0.1608 to 0.0675, and the point cloud coverage is improved from 76.64% to 98.95%.