Hancheng Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Hancheng Zhang has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Pengfei Liu
Institute of Highway Engineering, RWTH Aachen University
2 shared publications - Wei Zhou
School of Automation, Nanjing University of Science and Technology, Nanjing, China
2 shared publications Yuanyuan Hu
2 shared publicationsZhendong Qian
Intelligent Transportation System Research Center and the Bridge Engineering Research Center, Southeast University, Nanjing, China
2 shared publications- Chen Wang
Intelligent Transportation System Research Center, Southeast University, Nanjing 211189, China
1 shared publication Gang Liu
Intelligent Transportation System Research Center, Bridge Engineering Research Center, Southeast University, Nanjing, China
1 shared publication
Latest Publications
- Open Access
- 14 May 2025
- Published by Elsevier in Measurement
- Vol. 253
CrackAdaptNet: End-to-end domain adaptation for crack detection and quantification
- 14 May 2025
- Published by Elsevier in Measurement
- Vol. 253
Open AccessCrackAdaptNet: End-to-end domain adaptation for crack detection and quantification
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (12) , 20588-20601
Teaching Segment-Anything-Model Domain-Specific Knowledge for Road Crack Segmentation From On-Board Cameras
Road crack segmentation from on-board cameras is a highly desirable yet challenging task for road condition inspection and maintenance. However, existing methods trained on small-scale datasets present limited performance from such challenging perspectives due to the lack of sufficient prior knowledge and effective generalizability. To address this limitation, this paper incorporates a vision foundation model named Segment-Anything-Model (SAM) and fully leverages its rich prior knowledge and strong generalizability to achieve crack segmentation. Also, we construct a customized crack segmentation dataset shot from on-board cameras. Considering the direct use of SAM might not correctly segment cracks, some lightweight and learnable crack adaptation layers are developed and integrated into SAM’s image encoder, which take the patch embeddings of the input image and its high-frequency components within road regions as joint inputs. During training, the parameters of the crack adaptation layers are fine-tuned to acquire domain-specific knowledge, while the parameters of the image encoder remain frozen, preserving SAM’s rich prior knowledge. Additionally, a sparse prompt generation method is proposed based on the high-frequency components within road regions, which guides the SAM model to better focus on high-frequency regions that may contain cracks. Experimental results demonstrate that the proposed framework achieves state-of-the-art performance, with an improved average precision of 8.14% compared to Mask2Former. Furthermore, the parameter-efficient transfer learning framework significantly reduces the number of parameters requiring fine-tuning, thereby improving efficiency and reducing training costs. The dataset proposed in this paper is available at https://github.com/TRMetaGroup/CrackSeg.
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (12) , 20588-20601
Teaching Segment-Anything-Model Domain-Specific Knowledge for Road Crack Segmentation From On-Board Cameras
Road crack segmentation from on-board cameras is a highly desirable yet challenging task for road condition inspection and maintenance. However, existing methods trained on small-scale datasets present limited performance from such challenging perspectives due to the lack of sufficient prior knowledge and effective generalizability. To address this limitation, this paper incorporates a vision foundation model named Segment-Anything-Model (SAM) and fully leverages its rich prior knowledge and strong generalizability to achieve crack segmentation. Also, we construct a customized crack segmentation dataset shot from on-board cameras. Considering the direct use of SAM might not correctly segment cracks, some lightweight and learnable crack adaptation layers are developed and integrated into SAM’s image encoder, which take the patch embeddings of the input image and its high-frequency components within road regions as joint inputs. During training, the parameters of the crack adaptation layers are fine-tuned to acquire domain-specific knowledge, while the parameters of the image encoder remain frozen, preserving SAM’s rich prior knowledge. Additionally, a sparse prompt generation method is proposed based on the high-frequency components within road regions, which guides the SAM model to better focus on high-frequency regions that may contain cracks. Experimental results demonstrate that the proposed framework achieves state-of-the-art performance, with an improved average precision of 8.14% compared to Mask2Former. Furthermore, the parameter-efficient transfer learning framework significantly reduces the number of parameters requiring fine-tuning, thereby improving efficiency and reducing training costs. The dataset proposed in this paper is available at https://github.com/TRMetaGroup/CrackSeg.
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (11) , 19018-19029
Enhancing Rare Object Detection on Roadways Through Conditional Diffusion Models for Data Augmentation
The detection of rare objects within traffic environments is critically important for the safety and reliability of autonomous driving systems. However, the scarcity of sufficient data and the long-tail effect associated with rare objects pose significant challenges to their detection. To address this issue, this study introduces a Conditional Diffusion Model, the RareGenDiffuse Model, which generates images of rare objects by taking traffic background images and corresponding bounding box images as inputs. Based on diffusion models, the RareGenDiffuse Model enables control over the category and location of the generated rare objects by adjusting the bounding box images. Experimental results demonstrate that the proposed model offers the advantages of high clarity and controllability. Utilizing the generated rare object dataset significantly enhances the detection outcomes for rare objects. The methodology proposed in this study can effectively reduce the efforts required for data acquisition and labeling, presenting a valuable solution to the challenges of rare object detection in traffic scenarios.
- 12 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (11) , 19018-19029
Enhancing Rare Object Detection on Roadways Through Conditional Diffusion Models for Data Augmentation
The detection of rare objects within traffic environments is critically important for the safety and reliability of autonomous driving systems. However, the scarcity of sufficient data and the long-tail effect associated with rare objects pose significant challenges to their detection. To address this issue, this study introduces a Conditional Diffusion Model, the RareGenDiffuse Model, which generates images of rare objects by taking traffic background images and corresponding bounding box images as inputs. Based on diffusion models, the RareGenDiffuse Model enables control over the category and location of the generated rare objects by adjusting the bounding box images. Experimental results demonstrate that the proposed model offers the advantages of high clarity and controllability. Utilizing the generated rare object dataset significantly enhances the detection outcomes for rare objects. The methodology proposed in this study can effectively reduce the efforts required for data acquisition and labeling, presenting a valuable solution to the challenges of rare object detection in traffic scenarios.
- 27 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (8) , 8460-8472
Fine-Grained Vehicle Make and Model Recognition Framework Based on Magnetic Fingerprint
Fine-grained vehicle classification of make and model recognition is of practical significance and can be widely applied in the tasks such as traffic perception and control. To date, vision-based vehicle make and model recognition methods have improved through the years; however, they often fail in low-visibility scenes. This paper proposes a fine-grained vehicle model recognition framework based on magnetic fingerprints instead of videos from surveillance cameras. The magnetic fingerprint refers to the pattern of magnetic field data obtained when vehicles pass magnetic sensors, which have less sensitivity and more robustness towards low-visibility scenes than surveillance cameras. Specifically, the proposed framework first utilizes adversarial autoencoders to extract the magnetic fingerprint from the raw magnetic field data, keeping detailed information for fine-grained vehicle model recognition as complete as possible. Then, an AdaBoostSVM algorithm is used to identify vehicle models from the magnetic fingerprint and further alleviate the hard-negative issue to reduce false classifications. Massive experimental results show that the proposed fine-grained vehicle magnetic fingerprint/AdaBoostSVM method achieves state-of-the-art results on fine-grained vehicle make and model recognition. The average precision reached 94.3% and the average recall reached 94.2% in the dataset containing 36 vehicle models in total. Notably, the proposed method outperforms surveillance camera methods, especially in low-visibility scenes, making it a valuable addition to future intelligent transportation systems.
- 27 March 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 25 (8) , 8460-8472
Fine-Grained Vehicle Make and Model Recognition Framework Based on Magnetic Fingerprint
Fine-grained vehicle classification of make and model recognition is of practical significance and can be widely applied in the tasks such as traffic perception and control. To date, vision-based vehicle make and model recognition methods have improved through the years; however, they often fail in low-visibility scenes. This paper proposes a fine-grained vehicle model recognition framework based on magnetic fingerprints instead of videos from surveillance cameras. The magnetic fingerprint refers to the pattern of magnetic field data obtained when vehicles pass magnetic sensors, which have less sensitivity and more robustness towards low-visibility scenes than surveillance cameras. Specifically, the proposed framework first utilizes adversarial autoencoders to extract the magnetic fingerprint from the raw magnetic field data, keeping detailed information for fine-grained vehicle model recognition as complete as possible. Then, an AdaBoostSVM algorithm is used to identify vehicle models from the magnetic fingerprint and further alleviate the hard-negative issue to reduce false classifications. Massive experimental results show that the proposed fine-grained vehicle magnetic fingerprint/AdaBoostSVM method achieves state-of-the-art results on fine-grained vehicle make and model recognition. The average precision reached 94.3% and the average recall reached 94.2% in the dataset containing 36 vehicle models in total. Notably, the proposed method outperforms surveillance camera methods, especially in low-visibility scenes, making it a valuable addition to future intelligent transportation systems.
- 1 December 1997
- Published by Wiley in Journal of Texture Studies
- Vol. 28 (6) , 657-672
SEM IMAGE PROCESSING FOR FOOD STRUCTURE ANALYSIS
Image processing techniques were developed to characterize the cellular structure of puffed corn meal extrudates from their scanning electron microscope (SEM) images. Cross‐section SEM images were taken and digitized for product samples of different degrees of expansion. Two major approaches were developed to extract structure‐related image features. The air cells were also manually measured as the true characteristics of the cellular structure. Correlation and regression analyses were conducted to determine the effectiveness of the image features as measures of the structural characteristics. A number of image features were found to be good indicators of cell size and cell uniformity (R2‐values up to 0.95). The image processing techniques developed were useful in cellular structure characterization and they provide a rapid and consistent way for food cellular structure quantification and analysis.
- 1 December 1997
- Published by Wiley in Journal of Texture Studies
- Vol. 28 (6) , 657-672
SEM IMAGE PROCESSING FOR FOOD STRUCTURE ANALYSIS
Image processing techniques were developed to characterize the cellular structure of puffed corn meal extrudates from their scanning electron microscope (SEM) images. Cross‐section SEM images were taken and digitized for product samples of different degrees of expansion. Two major approaches were developed to extract structure‐related image features. The air cells were also manually measured as the true characteristics of the cellular structure. Correlation and regression analyses were conducted to determine the effectiveness of the image features as measures of the structural characteristics. A number of image features were found to be good indicators of cell size and cell uniformity (R2‐values up to 0.95). The image processing techniques developed were useful in cellular structure characterization and they provide a rapid and consistent way for food cellular structure quantification and analysis.