Jitong Kang

Mr. Jitong Kang


Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

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

Representative Publications

View All Publications
14 March 2024 in Sensors
Verified
Journal Article
Open access
PDF file
FYFan Yingbo
Fan Yingbo
SMShanjun Mao
Shanjun Mao
Mei LiMei Li
Mei Li
ZWZheng Wu
Zheng Wu
JKJitong Kang
Jitong Kang
With the continuous development of deep learning, the application of object detection based on deep neural networks in the coal mine has been expanding. Simultaneously, as the production applications demand higher recognition accuracy, most research chooses to enlarge the depth and parameters of the network to improve accuracy. However, due to the limited computing resources in the coal mining face, it is challenging to meet the computation demands of a large number of hardware resources. Therefore, this paper proposes a lightweight object detection algorithm designed specifically for the coal mining face, referred to as CM-YOLOv8. The algorithm introduces adaptive predefined anchor boxes tailored to the coal mining face dataset to enhance the detection performance of various targets. Simultaneously, a pruning method based on the L1 norm is designed, significantly compressing the model’s computation and parameter volume without compromising accuracy. The proposed algorithm is validated on the coal mining dataset DsLMF+, achieving a compression rate of 40% on the model volume with less than a 1% drop in accuracy. Comparative analysis with other existing algorithms demonstrates its efficiency and practicality in coal mining scenarios. The experiments confirm that CM-YOLOv8 significantly reduces the model’s computational requirements and volume while maintaining high accuracy.

25 December 2023 in Remote Sensing
Verified
Journal Article
Open access
PDF file
JKJitong Kang
Jitong Kang
Ning ChenNing Chen
Ning Chen
Mei LiMei Li
Mei Li
SMShanjun Mao
Shanjun Mao
HZHaoyuan Zhang
Haoyuan Zhang
Hui LiuHui Liu
Hui Liu
In recent years, point cloud segmentation technology has increasingly played a pivotal role in tunnel construction and maintenance. Currently, traditional methods for segmenting point clouds in tunnel scenes often rely on a multitude of attribute information, including spatial distribution, color, normal vectors, intensity, and density. However, the underground tunnel scenes show greater complexity than road tunnel scenes, such as dim light, indistinct boundaries of tunnel walls, and disordered pipelines. Furthermore, issues pertaining to data quality, such as the lack of color information and insufficient annotated data, contribute to the subpar performance of conventional point cloud segmentation algorithms. To address this issue, a 3D point cloud segmentation framework specifically for underground tunnels is proposed based on the Segment Anything Model (SAM). This framework effectively leverages the generalization capability of the visual foundation model to automatically adapt to various scenes and perform efficient segmentation of tunnel point clouds. Specifically, the tunnel is first sliced along its direction on the tunnel line. Then, each sliced point cloud is projected onto a two-dimensional plane. Various projection methods and point cloud coloring techniques are employed to enhance SAM’s segmentation performance in images. Finally, the semantic segmentation of the entire underground tunnel is achieved by a small set of manually annotated semantic labels used as prompts in a progressive and recursive manner. The key feature of this method lies in its independence from model training, as it directly and efficiently addresses tunnel point cloud segmentation challenges by capitalizing on the generalization capability of foundation model. Comparative experiments against classical region growing algorithms and PointNet++ deep learning algorithms demonstrate the superior performance of our proposed algorithm.

4 April 2024 in Sensors
Verified
Journal Article
Open access
PDF file
JKJitong Kang
Jitong Kang
Mei LiMei Li
Mei Li
SMShanjun Mao
Shanjun Mao
FYFan Yingbo
Fan Yingbo
ZWZheng Wu
Zheng Wu
BLBen Li
Ben Li
In recent years, the deformation detection technology for underground tunnels has played a crucial role in coal mine safety management. Currently, traditional methods such as the cross method and those employing the roof abscission layer monitoring instrument are primarily used for tunnel deformation detection in coal mines. With the advancement of photogrammetric methods, three-dimensional laser scanners have gradually become the primary method for deformation detection of coal mine tunnels. However, due to the high-risk confined spaces and distant distribution of coal mine tunnels, stationary three-dimensional laser scanning technology requires a significant amount of labor and time, posing certain operational risks. Currently, mobile laser scanning has become a popular method for coal mine tunnel deformation detection. This paper proposes a method for detecting point cloud deformation of underground coal mine tunnels based on a handheld three-dimensional laser scanner. This method utilizes SLAM laser radar to obtain complete point cloud information of the entire tunnel, while projecting the three-dimensional point cloud onto different planes to obtain the coordinates of the tunnel centerline. By using the calculated tunnel centerline, the three-dimensional point cloud data collected at different times are matched to the same coordinate system, and then the tunnel deformation parameters are analyzed separately from the global and cross-sectional perspectives. Through on-site collection of tunnel data, this paper verifies the feasibility of the algorithm and compares it with other centerline fitting and point cloud registration algorithms, demonstrating higher accuracy and meeting practical needs.

25 January 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
FYFan Yingbo
Fan Yingbo
SMShanjun Mao
Shanjun Mao
Mei LiMei Li
Mei Li
ZWZheng Wu
Zheng Wu
JKJitong Kang
Jitong Kang
With the continuous development of deep learning, the application of object detection based on deep neural network in the coal mine has been expanding. Simultaneously, as the production applications demand higher recognition accuracy, most research chooses to enlarge the depth and parameters of the network to improve accuracy. However, due to the limited computing resources in the coal mining face, it is challenging to meet the computation demands of a large number of hardware resources. Therefore, this paper proposes a lightweight object detection algorithm designed specifically for coal mining face, referred to as CM-YOLOv8. The algorithm introduces adaptive predefined anchor boxes tailored to the coal mining face dataset to enhance the detection performance of various targets. Simultaneously, a pruning method based on the L1 norm is designed, significantly compressing the model's computation and parameter volume without compromising accuracy. The proposed algorithm is validated on the coal mining dataset DsLMF+, achieving a compression rate of 40% on the model volume with less than a 1% drop in accuracy. Comparative analysis with other existing algorithms demonstrates its efficiency and practicality in coal mining scenarios. The experiments confirm that CM-YOLOv8 significantly reduces the model's computational requirements and volume while maintaining high accuracy.

28 February 2024 in Preprints.org
Verified
Preprint
Open access
PDF file
FYFan Yingbo
Fan Yingbo
SMShanjun Mao
Shanjun Mao
Mei LiMei Li
Mei Li
ZWZheng Wu
Zheng Wu
JKJitong Kang
Jitong Kang
BLBen Li
Ben Li
Image stitching is a crucial aspect of image processing. However, factors like perspective and environment often lead to irregular shapes in stitched images. Cropping or completion methods typically result in substantial loss of information. This paper proposes a method for rectifying irregularly images into rectangles using deformable meshes and residual networks. The method utilizes a convolutional neural network to quantify rigid structures of images. Choosing the most suitable mesh structure based on the extraction results, offering options such as triangular, rectangular, and hexagonal. Subsequently, the irregularly image, predefined mesh structure, and predicted mesh structure are input into a wide residual neural network for regression. The loss function comprises local and global, aimed at minimizing the loss of image information within the mesh and global target. This method not only significantly reduces information loss during rectification but also adapting to different images with various rigid structures. Validation on the DIR-D dataset shows this method outperforms state-of-the-art methods in image rectification.

29 August 2023 in Research Square
Verified
Preprint
Open access
PDF file
FYFan Yingbo
Fan Yingbo
SMShanjun Mao
Shanjun Mao
Mei LiMei Li
Mei Li
JKJitong Kang
Jitong Kang
BLBen Li
Ben Li
Image stitching is a fundamental pillar of computer vision, and its effectiveness hinges significantly on the quality of feature descriptors. However, the existing feature descriptors confront several challenges, including inadequate robustness to noise or rotational transformations and limited adaptability during hardware deployment. To address these limitations, this paper proposed a feature descriptor for image stitching, denoted as Lightweight Multi-Feature Descriptors (LMFD). By extensively extracting gradients, means, and global information surrounding the feature points, the feature descriptors are generated through various combinations to enhance the image stitching. This empowers the algorithm with formidable rotational invariance and noise resistance, thereby improving its accuracy and reliability. Furthermore, the feature descriptors take the form of binary matrices consisting of 0 and 1, which not only facilitates more efficient hardware deployment but also enhances computational efficiency. The utilization of binary matrices significantly reduces the computational complexity of the algorithm while preserving its efficacy. To validate the effectiveness of the LMFD, rigorous experimentation was conducted on the Hpatches and 2D-HeLa datasets. The results demonstrated that the LMFD outperformed state-of-the-art image matching algorithms in terms of accuracy. This empirical evidence solidifies the superiority of the LMFD and substantiates its potential for practical applications in various domains.