Author name cluster

Xinwen Zhou

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

2 author rows

JBHI Journal 2025 Journal Article

Utilizing Hybrid Mask and Upsampling Attention Gate for Multiple Immunohistochemistry Image Cell Recognition

Xinwen Zhou
Jingyuan Yang
Ke Cheng
Qiu Liu
Huizi Sha
Ran Wei
Jingting Jiang

Multi-immunohistochemistry (mIHC) is a crucial technique for simultaneous detection of multiple cellular phenotypes within a single tissue section. Its application in cancer diagnosis and treatment underscores the importance of developing reliable automated cell detection and classification methods for mIHC images. However, existing approaches face significant challenges due to high cell density, heterogeneity, and the laborious nature of annotation. This study presents a novel automated cell detection and classification model specifically designed to address these limitations. The proposed model leverages a simplified point-based annotation approach, significantly reducing annotation effort compared to conventional methods. A hybrid masking strategy combining Gaussian and circular masks is introduced to accurately capture the diverse morphological characteristics of different cell types. To enhance detail detection against complex backgrounds and robustness in highly heterogeneous environments, a novel Upsampling Attention Gate (UAG) is proposed. This module effectively improves feature extraction by focusing on relevant information within the image. Finally, a post-processing module is incorporated to address cell adhesion issues during detection, further enhancing the accuracy of the model. Extensive experiments on the mIHC dataset demonstrate that the proposed method achieves F1 scores of 0. 772 and 0. 747 for cell detection and classification, respectively, outperforming existing methods across various performance metrics. This study offers a promising solution to the challenges of automated cell detection and classification in mIHC images, paving the way for improved diagnosis and treatment in cancer research.

Details DOI

IROS Conference 2019 Conference Paper

ROI-based Robotic Grasp Detection for Object Overlapping Scenes

Hanbo Zhang
Xuguang Lan
Site Bai
Xinwen Zhou
Zhiqiang Tian
Nanning Zheng 0001

Grasp detection considering the affiliations between grasps and their owner in object overlapping scenes is a necessary and challenging task for the practical use of the robotic grasping approach. In this paper, a robotic grasp detection algorithm named ROI-GD is proposed to provide a feasible solution to this problem based on Region of Interest (ROI), which is the region proposal for objects. ROI-GD uses features from ROIs to detect grasps instead of the whole scene. It has two stages: the first stage is to provide ROIs in the input image and the second-stage is the grasp detector based on ROI features. We also contribute a multi-object grasp dataset, (a) which is much larger than Cornell Grasp Dataset, by labeling Visual Manipulation Relationship Dataset. Experimental results demonstrate that ROI-GD performs much better in object overlapping scenes and at the meantime, remains comparable with state-of-the-art grasp detection algorithms on Cornell Grasp Dataset and Jacquard Dataset. Robotic experiments demonstrate that ROI-GD can help robots grasp the target in single-object and multi-object scenes with the overall success rates of 92. 5% and 83. 8% respectively.

Details

IROS Conference 2018 Conference Paper

Fully Convolutional Grasp Detection Network with Oriented Anchor Box

Xinwen Zhou
Xuguang Lan
Hanbo Zhang
Zhiqiang Tian
Yang Zhang
Nanning Zheng 0001

In this paper, we present a real-time approach to predict multiple grasping poses for a parallel-plate robotic gripper using RGB images. A model with oriented anchor box mechanism is proposed and a new matching strategy is used during the training process. An end-to-end fully convolutional neural network is employed in our work. The network consists of two parts: the feature extractor and multi-grasp predictor. The feature extractor is a deep convolutional neural network. The multi-grasp predictor regresses grasp rectangles from predefined oriented rectangles, called oriented anchor boxes, and classifies the rectangles into graspable and ungraspable. On the standard Cornell Grasp Dataset, our model achieves an accuracy of 97. 74% and 96. 61% on image-wise split and object-wise split respectively, and outperforms the latest state-of-the-art approach by 1. 74% on image-wise split and 0. 51% on object-wise split.

Details