yuanze meng

Mr. yuanze meng


Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 March 2025 in Complex & Intelligent Systems
Verified
Journal Article
Open access
PDF file
WWWei Wang
Wei Wang
YMyuanze meng
yuanze meng
HLHan Li
Han Li
GCGuiyong Chang
Guiyong Chang
SLShun Li
Shun Li
CZChenghong Zhang
Chenghong Zhang
Convolutional neural networks (CNNs) are constrained in their capacity to model geometric transformations due to their fixed geometric structure. To overcome this problem, researchers introduce deformable convolution, which allows the convolution kernel to be deformable on the feature map. However, deformable convolution may introduce irrelevant contextual information during the learning process and thus affect the model performance. DCNv2 introduces a modulation mechanism to control the diffusion of the sampling points to control the degree of contribution of offsets through weights, but we find that such problems still exist in practical use. Therefore, we propose a new limit deformable convolution to address this problem, which enhances the model ability by adding adaptive limiting units to constrain the offsets and adjusts the weight constraints on the offsets to enhance the image-focusing ability. In the subsequent work, we perform lightweight work on the limit deformable convolution and design three kinds of LDBottleneck to adapt to different scenarios. The limit deformable network, equipped with the optimal LDBottleneck, demonstrated an improvement in mAP75 of 1.4% compared to DCNv1 and 1.1% compared to DCNv2 on the VOC2012+2007 dataset. Furthermore, on the CoCo2017 dataset, different backbones equipped with our limit deformable module achieved satisfactory results. The source code for this work is publicly available at https://github.com/1977245719/LDCN.

19 December 2025 in Pattern Recognition
Verified
Journal Article
Closed access
WWWei Wang
Wei Wang
WLWeimin Lei
Weimin Lei
WZWei Zhang
Wei Zhang
BSBojian Song
Bojian Song
GZGuanghui Zhang
Guanghui Zhang
YMyuanze meng
yuanze meng

15 March 2024 in Research Square
Verified
Preprint
Open access
PDF file
BYBo Yu
Bo Yu
SLShun Li
Shun Li
WWWei Wang
Wei Wang
YMyuanze meng
yuanze meng
DLDongying Liu
Dongying Liu
Aiming at the problems of traditional target detection algorithms such as high demand for model equipment, low detection accuracy, high leakage rate of overlapping targets, and repeated detection of the same target, we propose an improved multilayer deformable convolution as well as a method for fusing attention mechanisms, and use improved BiFPN feature pyramid feature fusion in the network constructed using this method. First, the improved deformable convolution is used to replace part of the ordinary convolution in the original network to enhance the ability of the modified network to extract feature targets from irregular targets; second, the improved cascade feature fusion module is used to enhance the ability of the neural network to represent features at different scales and semantic levels, and by using the improved cascade feature module, it makes the network able to extract features from feature maps at different levels or scales. features, followed by using the weighted bidirectional feature pyramid network connectivity module to stitch the feature maps of different layers along the channel dimension, which is used to enrich the network's representation capability of the target, and finally, the hybrid attention mechanism is added to the network, which reduces the interference of the background as well as non-target objects on the target, and improves the network's accuracy of target detection. The experimental results show that the improved model in this paper improves over the original model by 3.3\% and 4.7\% in mAP50 and mAP50-95.

2 February 2025 in Electronics
Verified
Journal Article
Open access
PDF file
WWWei Wang
Wei Wang
YMyuanze meng
yuanze meng
HLHan Li
Han Li
SLShun Li
Shun Li
CZChenghong Zhang
Chenghong Zhang
WLWeimin Lei
Weimin Lei
Model lightweighting is significant in edge computing and mobile devices. Current studies on fast network design mainly focuses on model computation compression and speedup. Many models aim to compress models by dealing with redundant feature maps. However, most of these methods choose to preserve the feature maps with simple manipulations and do not effectively reduce redundant feature maps. This paper proposes a new convolution module, PDConv, which compresses redundant feature maps to reduce network complexity and increase network width to maintain accuracy. PDConv (Partial Deep Convolution) outperforms traditional methods in handling redundant feature maps, particularly in deep networks. Its FLOPs are comparable to deep separable convolution but with higher accuracy. This paper proposes PDBottleNeck and PDC2f (Partial Deep CSPDarknet53 to 2-Stage FPN) and build the lightweight network PDNet for experimental validation using the PASCAL VOC dataset. Compared to the popular HorNet, our method achieves an improvement of more than 25% in FLOPs and 1.8% in mAP50:95 accuracy. On the CoCo2017 dataset, our large PDNet achieves a 0.5% improvement in mAP75 and lower FLOPs than the latest RepVit.