KAIYANG XU

Dr. KAIYANG XU

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

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

Representative Publications

View All Publications
15 March 2024 in Sensors
Verified
Journal Article
Open access
PDF file
KXKAIYANG XU
KAIYANG XU
Haibin WuHaibin Wu
Haibin Wu
Yuji IwahoriYuji Iwahori
Yuji Iwahori
XYXiaoyu Yu
Xiaoyu Yu
ZHZeyu Hu
Zeyu Hu
aili wangaili wang
aili wang
How to obtain internal cavity features and perform image matching is a great challenge for laparoscopic 3D reconstruction. This paper proposes a method for detecting and associating vascular features based on dual-branch weighted fusion vascular structure enhancement. Our proposed method is divided into three stages, including analyzing various types of minimally invasive surgery (MIS) images and designing a universal preprocessing framework to make our method generalized. We propose a Gaussian weighted fusion vascular structure enhancement algorithm using the dual-branch Frangi measure and MFAT (multiscale fractional anisotropic tensor) to address the structural measurement differences and uneven responses between venous vessels and microvessels, providing effective structural information for vascular feature extraction. We extract vascular features through dual-circle detection based on branch point characteristics, and introduce NMS (non-maximum suppression) to reduce feature point redundancy. We also calculate the ZSSD (zero sum of squared differences) and perform feature matching on the neighboring blocks of feature points extracted from the front and back frames. The experimental results show that the proposed method has an average accuracy and repeatability score of 0.7149 and 0.5612 in the Vivo data set, respectively. By evaluating the quantity, repeatability, and accuracy of feature detection, our method has more advantages and robustness than the existing methods.

9 April 2026 in Optics and Lasers in Engineering
Verified
Journal Article
Open access
KXKAIYANG XU
KAIYANG XU
Haibin WuHaibin Wu
Haibin Wu
YSYinghui Sun
Yinghui Sun
SWSen Wang
Sen Wang
ZHZeyu Hu
Zeyu Hu
AWAili Wang
Aili Wang

12 October 2022 in Embedded Selforganising Systems
Verified
Journal Article
Open access
PDF file
KXKAIYANG XU
KAIYANG XU
JWJingfang Wei
Jingfang Wei
ZHZeyu Hu
Zeyu Hu
JZJun Zhang
Jun Zhang
Haibin WuHaibin Wu
Haibin Wu
Yuji IwahoriYuji Iwahori
Yuji Iwahori
In this paper, the projection distortion correction method of patient's body surface is studied. First, the applicability of four typical projection image distortion correction methods to different surfaces were compared. Secondly, Based on the correction method based on quadratic surface fitting, and a correction method based on point cloud mapping was designed, which uses structured light to extract the point cloud information on the surface of the projection screen for coordinate transformation. Combined with the depth information, it solves the distortion problem of complex surface projection images, expands the surface applicability of the correction method, and improves the correction accuracy and real-time performance. The projection distortion correction of different types of surfaces is realized by simulation, and the correction effect of the two methods is compared. The correction time is shortened from 67s to 7s by the correction method based on point cloud mapping. Finally, organ simulation experiment was used to verify the intraoperative feasibility, and the virtual "perspective" display effect was presented as a whole.

11 January 2026 in Electronics
Verified
Journal Article
Open access
PDF file
KXKAIYANG XU
KAIYANG XU
Haibin WuHaibin Wu
Haibin Wu
Liang YuLiang Yu
Liang Yu
XHXin He
Xin He
To address the application requirements of Visual Simultaneous Localization and Mapping (VSLAM) in intracavitary environments and the scarcity of gold-standard datasets for deep learning methods, this study proposes a hybrid attention mechanism combined with U-Net for vascular branch point extraction in endoluminal images (SuperVessel). The network is initialized via transfer learning with pre-trained SuperRetina model parameters and integrated with a vascular feature detection and matching method based on dual branch fusion and structure enhancement, generating a pseudo-gold-standard vascular branch point dataset. The framework employs a dual-decoder architecture, incorporates a dynamic up-sampling module (CBAM-Dysample) to refine local vessel features through hybrid attention mechanisms, designs a Dice-Det loss function weighted by branching features to prioritize vessel junctions, and introduces a dynamically weighted Triplet-Des loss function optimized for descriptor discrimination. Experiments on the Vivo test set demonstrate that the proposed method achieves an average Area Under Curve (AUC) of 0.760, with mean feature points, accuracy, and repeatability scores of 42,795, 0.5294, and 0.46, respectively. Compared to SuperRetina, the method maintains matching stability while exhibiting superior repeatability, feature point density, and robustness in low-texture/deformation scenarios. Ablation studies confirm the CBAM-Dysample module’s efficacy in enhancing feature expression and convergence speed, offering a robust solution for intracavitary SLAM systems.

24 November 2023
Verified
Conference Paper
Open access
ZHZeyu Hu
Zeyu Hu
HCHaisong Chen
Haisong Chen
JWJingfang Wei
Jingfang Wei
YZYunyi Zuo
Yunyi Zuo
KXKAIYANG XU
KAIYANG XU
aili wangaili wang
aili wang
With the rapid development of artificial intelligence, the emergence of intelligent vehicles has brought disruptive changes to traditional survey technology. In response to the limitations of geological environmental factors on the scope and search progress of field exploration, this article utilizes a modular design concept to develop a system structure framework and specific solutions for each functional module of a multi-functional intelligent survey vehicle. An intelligent control platform based on the Nvidia Jetson TX2 and ROS systems has been built, and software programs for eye movement control and 3D dense reconstruction have been written. A system that can assist motion control through line of sight has been designed, A small, low-cost intelligent survey vehicle that can simultaneously perform real-time scene 3D reconstruction. This survey vehicle combines IoT technology, and on the basis of traditional survey vehicle functions, it has functions such as eye tracking, scene 3D reconstruction, remote object detection, etc. The obtained information is transmitted back to the PC through remote communication for users to view the survey results in real-time. At the same time, the operation of the survey vehicle can be controlled according to the direction of the line of sight, providing intelligent interaction ability and multi-dimensional exploration results for exploration work.

7 January 2026 in Results in Engineering
Verified
Journal Article
Open access
Haibin WuHaibin Wu
Haibin Wu
KXKAIYANG XU
KAIYANG XU
ZHZeyu Hu
Zeyu Hu
aili wangaili wang
aili wang