Qin Qin
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Qin Qin has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
Zimei Tu
School of Intelligent Manufacturing and Control Engineering, Shanghai Polytechnic University, Shanghai, 201209, China
4 shared publications- Dongfang Zhao
School of Intelligent Manufacturing and Control Engineering, Shanghai Polytechnic University, Shanghai, China
2 shared publications Jianfang Dou
Department of Automation and Mechanical and Electrical engineering, School of Intelligent Manufacturing and Control Engineering, Shanghai Polytechnic University, 201209, China
2 shared publications- Jianfang Dou
Department of Automation and Mechanical and Electrical engineering, School of Intelligent Manufacturing and Control Engineering, Shanghai Second Polytechnic University, Shanghai, China
2 shared publications Shulin Liu
School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, China
2 shared publications- Z. Tu
Department of Automation and Mechanical and Electrical engineering, School of Intelligent Manufacturing and Control Engineering, Shanghai Second Polytechnic University, Shanghai, China
2 shared publications
Latest Publications
- 17 September 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (12) , 649
Generalized composite symbolic aggregate approximation and laplacian score based feature extraction scheme for mechanical fault diagnosis
- 17 September 2026
- Published by Springer Nature in Signal, Image and Video Processing
- Vol. 20 (12) , 649
Generalized composite symbolic aggregate approximation and laplacian score based feature extraction scheme for mechanical fault diagnosis
- 1 July 2026
- Published by AIP Publishing in Review of Scientific Instruments
- Vol. 97 (7)
Compressed multi-scale entropy and its application in mechanical fault diagnosis
Multi-scale entropy (MSEn) is a powerful tool for measuring the complexity of time series, and it has been widely recognized in fault diagnosis. However, most of the current multi-scale entropy methods ignore information aliasing caused by the coarse-graining process, which makes it difficult to maintain the information reliability of the original data and thus affects the analysis results. In response to the aforementioned issue, this paper first analyzes the information aliasing of the classical MSEn and then proposes the compressed multi-scale entropy (CoMSEn) for mechanical fault diagnosis. In the proposed CoMSEn, the random Gaussian matrix is introduced to achieve transformation domain projection, and the results under different compression rates are treated as the coarse-graining results under different scales to achieve multi-scale analysis. Furthermore, to eliminate the influence of randomness of the Gaussian matrix, the mean value of the coarse-grained sequence sample entropy obtained through multiple compressions is adopted as the final result. Benefiting from the random Gaussian matrix satisfying the restricted isometric property with high probability under the assumption that the original signal is sparse in the specific transform domain, the coarse-grained sequences obtained under different compression rates can preserve the essential discriminative information of the raw data and effectively suppress the information aliasing caused by the linear down-sampling in classical MSEn, thereby providing a theoretical basis for the entropy calculation results’ reliability. The effectiveness of the proposed method is verified utilizing the CWRU rolling bearing dataset and an actual reciprocating compressor valve dataset.
- 1 July 2026
- Published by AIP Publishing in Review of Scientific Instruments
- Vol. 97 (7)
Compressed multi-scale entropy and its application in mechanical fault diagnosis
Multi-scale entropy (MSEn) is a powerful tool for measuring the complexity of time series, and it has been widely recognized in fault diagnosis. However, most of the current multi-scale entropy methods ignore information aliasing caused by the coarse-graining process, which makes it difficult to maintain the information reliability of the original data and thus affects the analysis results. In response to the aforementioned issue, this paper first analyzes the information aliasing of the classical MSEn and then proposes the compressed multi-scale entropy (CoMSEn) for mechanical fault diagnosis. In the proposed CoMSEn, the random Gaussian matrix is introduced to achieve transformation domain projection, and the results under different compression rates are treated as the coarse-graining results under different scales to achieve multi-scale analysis. Furthermore, to eliminate the influence of randomness of the Gaussian matrix, the mean value of the coarse-grained sequence sample entropy obtained through multiple compressions is adopted as the final result. Benefiting from the random Gaussian matrix satisfying the restricted isometric property with high probability under the assumption that the original signal is sparse in the specific transform domain, the coarse-grained sequences obtained under different compression rates can preserve the essential discriminative information of the raw data and effectively suppress the information aliasing caused by the linear down-sampling in classical MSEn, thereby providing a theoretical basis for the entropy calculation results’ reliability. The effectiveness of the proposed method is verified utilizing the CWRU rolling bearing dataset and an actual reciprocating compressor valve dataset.
- Open Access
- 1 January 2025
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 83 (2) , 3211-3226
Deep Learning Based Online Defect Detection Method for Automotive Sealing Rings
Manufacturers must identify and classify various defects in automotive sealing rings to ensure product quality. Deep learning algorithms show promise in this field, but challenges remain, especially in detecting small-scale defects under harsh industrial conditions with multimodal data. This paper proposes an enhanced version of You Only Look Once (YOLO)v8 for improved defect detection in automotive sealing rings. We introduce the Multi-scale Adaptive Feature Extraction (MAFE) module, which integrates Deformable Convolutional Network (DCN) and Space-to-Depth (SPD) operations. This module effectively captures long-range dependencies, enhances spatial aggregation, and minimizes information loss of small objects during feature extraction. Furthermore, we introduce the Blur-Aware Wasserstein Distance (BAWD) loss function, which improves regression accuracy and detection capabilities for small object anchor boxes, particularly in scenarios involving defocus blur. Additionally, we have constructed a high-quality dataset of automotive sealing ring defects, providing a valuable resource for evaluating defect detection methods. Experimental results demonstrate our method’s high performance, achieving 98.30% precision, 96.62% recall, and an inference speed of 20.3 ms.
- 1 January 2025
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 83 (2) , 3211-3226
Open AccessDeep Learning Based Online Defect Detection Method for Automotive Sealing Rings
Manufacturers must identify and classify various defects in automotive sealing rings to ensure product quality. Deep learning algorithms show promise in this field, but challenges remain, especially in detecting small-scale defects under harsh industrial conditions with multimodal data. This paper proposes an enhanced version of You Only Look Once (YOLO)v8 for improved defect detection in automotive sealing rings. We introduce the Multi-scale Adaptive Feature Extraction (MAFE) module, which integrates Deformable Convolutional Network (DCN) and Space-to-Depth (SPD) operations. This module effectively captures long-range dependencies, enhances spatial aggregation, and minimizes information loss of small objects during feature extraction. Furthermore, we introduce the Blur-Aware Wasserstein Distance (BAWD) loss function, which improves regression accuracy and detection capabilities for small object anchor boxes, particularly in scenarios involving defocus blur. Additionally, we have constructed a high-quality dataset of automotive sealing ring defects, providing a valuable resource for evaluating defect detection methods. Experimental results demonstrate our method’s high performance, achieving 98.30% precision, 96.62% recall, and an inference speed of 20.3 ms.
- 1 January 2021
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 31 (1) , 7-17
Multi-Modal Image Registration Based on Local Self-Similarity and Bidirectional Matching
The registration of images from infrared sensors and visible color sensors is a quite difficult problem due to their different phenomena. In this paper, we propose a new method to register visible and infrared images. The proposed approach consists of three main steps. In the first step, SURF (Speeded-Up Robust Features) algorithm is applied for local feature extraction. In the second step, LSS (local self-similarity descriptor) is computed for each extracted feature. Finally, a cross matching process followed by a consistency check in the projective transformation model is performed for feature correspondence and mismatch elimination. Experimental results show the proposed method achieves better accuracy for registering visible and infrared images as compared to state-of-the-art approaches.
- 1 January 2021
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 31 (1) , 7-17
Multi-Modal Image Registration Based on Local Self-Similarity and Bidirectional Matching
The registration of images from infrared sensors and visible color sensors is a quite difficult problem due to their different phenomena. In this paper, we propose a new method to register visible and infrared images. The proposed approach consists of three main steps. In the first step, SURF (Speeded-Up Robust Features) algorithm is applied for local feature extraction. In the second step, LSS (local self-similarity descriptor) is computed for each extracted feature. Finally, a cross matching process followed by a consistency check in the projective transformation model is performed for feature correspondence and mismatch elimination. Experimental results show the proposed method achieves better accuracy for registering visible and infrared images as compared to state-of-the-art approaches.
- 1 July 2020
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 30 (3) , 541-550
Deep ResNet Based Remote Sensing Image Super-Resolution Reconstruction in Discrete Wavelet Domain
We present a single-image super-resolution (SR) method for Remote Sensing Image based on deep learning within Discrete Wavelet Domain in this paper. Our method is inspired Residual Learning. Firstly, an input image is decomposed by single level 2D Discrete wavelet transform to get four sub-bands. The four sub-bands coefficients are feeding into the Deep Learning Residual Network to predict correspondingly residual images; Adding four sub-band images and residual images as the new sub-bands of 2D wavelet transform; Finally, uses the inverse 2D Discrete wavelet transform to get the final output Super Resolution HR image. Our proposed method performs better than existing methods in accuracy and visual improvements in our results are easily noticeable.
- 1 July 2020
- Published by Pleiades Publishing Ltd in Pattern Recognition and Image Analysis
- Vol. 30 (3) , 541-550
Deep ResNet Based Remote Sensing Image Super-Resolution Reconstruction in Discrete Wavelet Domain
We present a single-image super-resolution (SR) method for Remote Sensing Image based on deep learning within Discrete Wavelet Domain in this paper. Our method is inspired Residual Learning. Firstly, an input image is decomposed by single level 2D Discrete wavelet transform to get four sub-bands. The four sub-bands coefficients are feeding into the Deep Learning Residual Network to predict correspondingly residual images; Adding four sub-band images and residual images as the new sub-bands of 2D wavelet transform; Finally, uses the inverse 2D Discrete wavelet transform to get the final output Super Resolution HR image. Our proposed method performs better than existing methods in accuracy and visual improvements in our results are easily noticeable.
- 2 April 2020
- Published by Springer Nature
- p. 576-585
The Neutrophil’s Morphology Classification Using Convolutional Neural Network
- 2 April 2020
- Published by Springer Nature
- p. 576-585
The Neutrophil’s Morphology Classification Using Convolutional Neural Network
- 15 December 2018
- Published by Springer Nature
- p. 411-417
Health Detection System for Skyscrapers
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5819-5822
Hierarchical image matching algorithm based on SIFT
In this paper, an algorithm for a robust template matching method based on the combination of the wavelet transform method and scale-invariant feature transform (SIFT) is proposed. Discrete wavelet transform is done firstly on a reference image and a template image, and low frequency parts of them are extracted, then we use Harris corner detection to detect the interesting point in low frequency parts of them to determine the matching candidate region of template image in reference image, extracting SIFT features on the matching candidate region and template image, The extracted SIFT features are matched by k-d tree and bidirectional matching strategy, then we exploited the information from SIFT to comprise matching constraint and use them to get more correct matches. Experiment show that, the algorithm can improve the accuracy of matching.
- 1 June 2018
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5819-5822
Hierarchical image matching algorithm based on SIFT
In this paper, an algorithm for a robust template matching method based on the combination of the wavelet transform method and scale-invariant feature transform (SIFT) is proposed. Discrete wavelet transform is done firstly on a reference image and a template image, and low frequency parts of them are extracted, then we use Harris corner detection to detect the interesting point in low frequency parts of them to determine the matching candidate region of template image in reference image, extracting SIFT features on the matching candidate region and template image, The extracted SIFT features are matched by k-d tree and bidirectional matching strategy, then we exploited the information from SIFT to comprise matching constraint and use them to get more correct matches. Experiment show that, the algorithm can improve the accuracy of matching.
- 11 February 2018
- Published by Springer Nature
- p. 591-597
Sorting System of Robot Based on Vision Detection
- 11 February 2018
- Published by Springer Nature
- p. 591-597
Sorting System of Robot Based on Vision Detection