Changzhen Qiu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Changzhen Qiu has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Zhiyong Zhang
School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen, China
10 shared publications - Shaolong Chen
School of Sino-German Intelligent Manufacturing, Shenzhen City Polytechnic, Shenzhen, China
3 shared publications Shaolong Chen
School of Electronics and Communication Engineering, Sun Yat-sen University, Guangzhou, China
2 shared publications- Weiping Yang
Sun Yat-Sen University, Shenzhen, CHINA
2 shared publications - Xintao Chen
School of Electronics and Communication Engineering, Sun Yat-sen University, Shenzhen 518107, China
2 shared publications Yuhan Xie
School of Electronics and Communication Engineering, Sun Yat-sen University, Guangzhou, China
2 shared publications
Latest Publications
- 19 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (17) , 28338-28350
AGMFusion: A Real-Time End-to-End Infrared and Visible Image Fusion Network Based on Adaptive Guidance Module
The goal of infrared and visible image fusion is to combine complementary information from the source images. However, owing to the absence of ground truth, most fusion algorithms just utilize information from source images, which cannot provide targeted guidance for network learning, leading to suboptimal outcomes. Moreover, numerous methods focus solely on modifying network architecture to improve fusion performance, without optimizing fusion algorithms from other perspectives. To tackle these problems, we propose a real-time end-to-end infrared and visible image fusion network based on adaptive guidance module (AGMFusion). On the one hand, by combining state-of-the-art methods and best fusion outcomes during the training process, the adaptive guidance module effectively facilitates collaborative guidance for network training. Additionally, based on the adaptive guidance module, we devise a loss function that includes content loss and guidance loss. We further balance these two loss components with the adaptive weight to boost the performance of our framework. On the other hand, AGMFusion is a lightweight image fusion framework that is able to generate highly perceptual fused images while maintaining excellent real-time performance, which can be potentially deployed as a pre-processing unit for various vision tasks. According to extensive comparative and generalization experiments, AGMFusion exceeds existing methods in visual appeal and quantitative metrics. Importantly, performance comparisons of numerous frameworks in running efficiency and object detection highlight the advantages of our approach. The source code will be released at https://github.com/liushh39/AGMFusion.
- 19 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 24 (17) , 28338-28350
AGMFusion: A Real-Time End-to-End Infrared and Visible Image Fusion Network Based on Adaptive Guidance Module
The goal of infrared and visible image fusion is to combine complementary information from the source images. However, owing to the absence of ground truth, most fusion algorithms just utilize information from source images, which cannot provide targeted guidance for network learning, leading to suboptimal outcomes. Moreover, numerous methods focus solely on modifying network architecture to improve fusion performance, without optimizing fusion algorithms from other perspectives. To tackle these problems, we propose a real-time end-to-end infrared and visible image fusion network based on adaptive guidance module (AGMFusion). On the one hand, by combining state-of-the-art methods and best fusion outcomes during the training process, the adaptive guidance module effectively facilitates collaborative guidance for network training. Additionally, based on the adaptive guidance module, we devise a loss function that includes content loss and guidance loss. We further balance these two loss components with the adaptive weight to boost the performance of our framework. On the other hand, AGMFusion is a lightweight image fusion framework that is able to generate highly perceptual fused images while maintaining excellent real-time performance, which can be potentially deployed as a pre-processing unit for various vision tasks. According to extensive comparative and generalization experiments, AGMFusion exceeds existing methods in visual appeal and quantitative metrics. Importantly, performance comparisons of numerous frameworks in running efficiency and object detection highlight the advantages of our approach. The source code will be released at https://github.com/liushh39/AGMFusion.
- 17 July 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (1) , 191-202
Fully and Weakly Supervised Deep Learning for Meniscal Injury Classification, and Location Based on MRI
Meniscal injury is a common cause of knee joint pain and a precursor to knee osteoarthritis (KOA). The purpose of this study is to develop an automatic pipeline for meniscal injury classification and localization using fully and weakly supervised networks based on MRI images. In this retrospective study, data were from the osteoarthritis initiative (OAI). The MR images were reconstructed using a sagittal intermediate-weighted fat-suppressed turbo spin-echo sequence. (1) We used 130 knees from the OAI to develop the LGSA-UNet model which fuses the features of adjacent slices and adjusts the blocks in Siam to enable the central slice to obtain rich contextual information. (2) One thousand seven hundred and fifty-six knees from the OAI were included to establish segmentation and classification models. The segmentation model achieved a DICE coefficient ranging from 0.84 to 0.93. The AUC values ranged from 0.85 to 0.95 in the binary models. The accuracy for the three types of menisci (normal, tear, and maceration) ranged from 0.60 to 0.88. Furthermore, 206 knees from the orthopedic hospital were used as an external validation data set to evaluate the performance of the model. The segmentation and classification models still performed well on the external validation set. To compare the diagnostic performances between the deep learning (DL) models and radiologists, the external validation sets were sent to two radiologists. The binary classification model outperformed the diagnostic performance of the junior radiologist (0.82–0.87 versus 0.74–0.88). This study highlights the potential of DL in knee meniscus segmentation and injury classification which can help improve diagnostic efficiency.
- 17 July 2024
- Published by Springer Nature in Journal of Imaging Informatics in Medicine
- Vol. 38 (1) , 191-202
Fully and Weakly Supervised Deep Learning for Meniscal Injury Classification, and Location Based on MRI
Meniscal injury is a common cause of knee joint pain and a precursor to knee osteoarthritis (KOA). The purpose of this study is to develop an automatic pipeline for meniscal injury classification and localization using fully and weakly supervised networks based on MRI images. In this retrospective study, data were from the osteoarthritis initiative (OAI). The MR images were reconstructed using a sagittal intermediate-weighted fat-suppressed turbo spin-echo sequence. (1) We used 130 knees from the OAI to develop the LGSA-UNet model which fuses the features of adjacent slices and adjusts the blocks in Siam to enable the central slice to obtain rich contextual information. (2) One thousand seven hundred and fifty-six knees from the OAI were included to establish segmentation and classification models. The segmentation model achieved a DICE coefficient ranging from 0.84 to 0.93. The AUC values ranged from 0.85 to 0.95 in the binary models. The accuracy for the three types of menisci (normal, tear, and maceration) ranged from 0.60 to 0.88. Furthermore, 206 knees from the orthopedic hospital were used as an external validation data set to evaluate the performance of the model. The segmentation and classification models still performed well on the external validation set. To compare the diagnostic performances between the deep learning (DL) models and radiologists, the external validation sets were sent to two radiologists. The binary classification model outperformed the diagnostic performance of the junior radiologist (0.82–0.87 versus 0.74–0.88). This study highlights the potential of DL in knee meniscus segmentation and injury classification which can help improve diagnostic efficiency.
- Open Access
A Multiscale Method for Infrared Ship Detection Based on Morphological Reconstruction and Two-Branch Compensation Strategy
Infrared ship target detection is crucial technology in marine scenarios. Ship targets vary in scale throughout navigation because the distance between the ship and the infrared camera is constantly changing. Furthermore, complex backgrounds, such as sea clutter, can cause significant interference during detection tasks. In this paper, multiscale morphological reconstruction-based saliency mapping, combined with a two-branch compensation strategy (MMRSM-TBC) algorithm, is proposed for the detection of ship targets of various sizes and against complex backgrounds. First, a multiscale morphological reconstruction method is proposed to enhance the ship targets in the infrared image and suppress any irrelevant background. Then, by introducing a structure tensor with two feature-based filter templates, we utilize the contour information of the ship targets and further improve their intensities in the saliency map. After that, a two-branch compensation strategy is proposed, due to the uneven distribution of image grayscale. Finally, the target is extracted using an adaptive threshold. The experimental results fully show that our proposed algorithm achieves strong performance in the detection of different-sized ship targets and has a higher accuracy than other existing methods.
Open AccessA Multiscale Method for Infrared Ship Detection Based on Morphological Reconstruction and Two-Branch Compensation Strategy
Infrared ship target detection is crucial technology in marine scenarios. Ship targets vary in scale throughout navigation because the distance between the ship and the infrared camera is constantly changing. Furthermore, complex backgrounds, such as sea clutter, can cause significant interference during detection tasks. In this paper, multiscale morphological reconstruction-based saliency mapping, combined with a two-branch compensation strategy (MMRSM-TBC) algorithm, is proposed for the detection of ship targets of various sizes and against complex backgrounds. First, a multiscale morphological reconstruction method is proposed to enhance the ship targets in the infrared image and suppress any irrelevant background. Then, by introducing a structure tensor with two feature-based filter templates, we utilize the contour information of the ship targets and further improve their intensities in the saliency map. After that, a two-branch compensation strategy is proposed, due to the uneven distribution of image grayscale. Finally, the target is extracted using an adaptive threshold. The experimental results fully show that our proposed algorithm achieves strong performance in the detection of different-sized ship targets and has a higher accuracy than other existing methods.
- Open Access
- 10 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 98-104
An atrium segmentation network with location guidance and siamese adjustment
- 10 February 2023
- Published by Association for Computing Machinery (ACM)
- p. 98-104
Open AccessAn atrium segmentation network with location guidance and siamese adjustment
- Open Access
An atrium segmentation network with location guidance and siamese adjustment
The segmentation of atrial scan images is of great significance for the three-dimensional reconstruction of the atrium and the surgical positioning. Most of the existing segmentation networks adopt a 2D structure and only take original images as input, ignoring the context information of 3D images and the role of prior information. In this paper, we propose an atrium segmentation network LGSANet with location guidance and siamese adjustment, which takes adjacent three slices of images as input and adopts an end-to-end approach to achieve coarse-to-fine atrial segmentation. The location guidance(LG) block uses the prior information of the localization map to guide the encoding features of the fine segmentation stage, and the siamese adjustment(SA) block uses the context information to adjust the segmentation edges. On the atrium datasets of ACDC and ASC, sufficient experiments prove that our method can adapt to many classic 2D segmentation networks, so that it can obtain significant performance improvements.
Open AccessAn atrium segmentation network with location guidance and siamese adjustment
The segmentation of atrial scan images is of great significance for the three-dimensional reconstruction of the atrium and the surgical positioning. Most of the existing segmentation networks adopt a 2D structure and only take original images as input, ignoring the context information of 3D images and the role of prior information. In this paper, we propose an atrium segmentation network LGSANet with location guidance and siamese adjustment, which takes adjacent three slices of images as input and adopts an end-to-end approach to achieve coarse-to-fine atrial segmentation. The location guidance(LG) block uses the prior information of the localization map to guide the encoding features of the fine segmentation stage, and the siamese adjustment(SA) block uses the context information to adjust the segmentation edges. On the atrium datasets of ACDC and ASC, sufficient experiments prove that our method can adapt to many classic 2D segmentation networks, so that it can obtain significant performance improvements.
- RetractedOpen Access
- 1 January 2023
- Published by Wiley in Journal of Sensors
- Vol. 2023 (1)
[Retracted] FMT Selector: Fourier‐Mellin Transformer with High Speed Rotating in Ship Target Detection and Tracking Based on Internet of Things and Wireless Sensor Network
The terminal guidance tracking always has high speed rotating and scale changing. It makes a large margin of error in current tracking algorithms, which can transmit wrong position in the wireless sensor networks of the Internet of Things. In this paper, we propose a selector based on Fourier‐Mellin transformer, which can accurately track the target with high speed rotating. Firstly we use existing detector to get all proposal target per frame. Then, we use Fourier‐Mellin matching to select the best target in the new frame. Our dataset is the ship video which is created by HRSC2016 in rotating. Experimental results show that the accuracy of this method is up to 89.8%. Compared with other traditional tracking algorithms, it has a leap forward. The FMT selector can be applied to the back end of the detector, and it can achieve a feasible effect in the field of terminal guidance tracking.
- 1 January 2023
- Published by Wiley in Journal of Sensors
- Vol. 2023 (1)
RetractedOpen Access[Retracted] FMT Selector: Fourier‐Mellin Transformer with High Speed Rotating in Ship Target Detection and Tracking Based on Internet of Things and Wireless Sensor Network
The terminal guidance tracking always has high speed rotating and scale changing. It makes a large margin of error in current tracking algorithms, which can transmit wrong position in the wireless sensor networks of the Internet of Things. In this paper, we propose a selector based on Fourier‐Mellin transformer, which can accurately track the target with high speed rotating. Firstly we use existing detector to get all proposal target per frame. Then, we use Fourier‐Mellin matching to select the best target in the new frame. Our dataset is the ship video which is created by HRSC2016 in rotating. Experimental results show that the accuracy of this method is up to 89.8%. Compared with other traditional tracking algorithms, it has a leap forward. The FMT selector can be applied to the back end of the detector, and it can achieve a feasible effect in the field of terminal guidance tracking.
- Open Access
- 28 November 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (23) , 12181
Sequence-to-Sequence Multi-Agent Reinforcement Learning for Multi-UAV Task Planning in 3D Dynamic Environment
Task planning involving multiple unmanned aerial vehicles (UAVs) is one of the main research topics in the field of cooperative unmanned aerial vehicle control systems. This is a complex optimization problem where task allocation and path planning are dealt with separately. However, the recalculation of optimal results is too slow for real-time operations in dynamic environments due to a large amount of computation required, and traditional algorithms are difficult to handle scenarios of varying scales. Meanwhile, the traditional approach confines task planning to a 2D environment, which deviates from the real world. In this paper, we design a 3D dynamic environment and propose a method for task planning based on sequence-to-sequence multi-agent deep deterministic policy gradient (SMADDPG) algorithm. First, we construct the task-planning problem as a multi-agent system based on the Markov decision process. Then, the DDPG is combined sequence-to-sequence to learn the system to solve task assignment and path planning simultaneously according to the corresponding reward function. We compare our approach with the traditional reinforcement learning algorithm in this system. The simulation results show that our approach satisfies the task-planning requirements and can accomplish tasks more efficiently in competitive as well as cooperative scenarios with dynamic or constant scales.
- 28 November 2022
- Published by MDPI AG in Applied Sciences
- Vol. 12 (23) , 12181
Open AccessSequence-to-Sequence Multi-Agent Reinforcement Learning for Multi-UAV Task Planning in 3D Dynamic Environment
Task planning involving multiple unmanned aerial vehicles (UAVs) is one of the main research topics in the field of cooperative unmanned aerial vehicle control systems. This is a complex optimization problem where task allocation and path planning are dealt with separately. However, the recalculation of optimal results is too slow for real-time operations in dynamic environments due to a large amount of computation required, and traditional algorithms are difficult to handle scenarios of varying scales. Meanwhile, the traditional approach confines task planning to a 2D environment, which deviates from the real world. In this paper, we design a 3D dynamic environment and propose a method for task planning based on sequence-to-sequence multi-agent deep deterministic policy gradient (SMADDPG) algorithm. First, we construct the task-planning problem as a multi-agent system based on the Markov decision process. Then, the DDPG is combined sequence-to-sequence to learn the system to solve task assignment and path planning simultaneously according to the corresponding reward function. We compare our approach with the traditional reinforcement learning algorithm in this system. The simulation results show that our approach satisfies the task-planning requirements and can accomplish tasks more efficiently in competitive as well as cooperative scenarios with dynamic or constant scales.
- Open Access
- 31 October 2022
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2022, 1-10
Multiresolution Mutual Assistance Network for Cardiac Magnetic Resonance Images Segmentation
The automatic segmentation of cardiac magnetic resonance (MR) images is the basis for the diagnosis of cardiac-related diseases. However, the segmentation of cardiac MR images is a challenging task due to the inhomogeneity of MR images intensity distribution and the unclear boundaries between adjacent tissues. In this paper, we propose a novel multiresolution mutual assistance network (MMA-Net) for cardiac MR images segmentation. It is mainly composed of multibranch input module, multiresolution mutual assistance module, and multilabel deep supervision. First, the multibranch input module helps the network to extract local and global features more pertinently. Then, the multiresolution mutual assistance module implements multiresolution feature interaction and progressively improves semantic features to more completely express the information of the tissue. Finally, the multilabel deep supervision is proposed to generate the final segmentation map. We compare with state-of-the-art medical image segmentation methods on the medical image computing and computer-assisted intervention (MICCAI) automated cardiac diagnosis challenge datasets and the MICCAI atrial segmentation challenge datasets. The mean dice scores of our method in the left atrium, right ventricle, myocardium, and left ventricle are 0.919, 0.920, 0.881, and 0.960, respectively. The analysis of evaluation indicators and segmentation results shows that our method achieves the best performance in cardiac magnetic resonance images segmentation.
- 31 October 2022
- Published by Wiley in Journal of Healthcare Engineering
- Vol. 2022, 1-10
Open AccessMultiresolution Mutual Assistance Network for Cardiac Magnetic Resonance Images Segmentation
The automatic segmentation of cardiac magnetic resonance (MR) images is the basis for the diagnosis of cardiac-related diseases. However, the segmentation of cardiac MR images is a challenging task due to the inhomogeneity of MR images intensity distribution and the unclear boundaries between adjacent tissues. In this paper, we propose a novel multiresolution mutual assistance network (MMA-Net) for cardiac MR images segmentation. It is mainly composed of multibranch input module, multiresolution mutual assistance module, and multilabel deep supervision. First, the multibranch input module helps the network to extract local and global features more pertinently. Then, the multiresolution mutual assistance module implements multiresolution feature interaction and progressively improves semantic features to more completely express the information of the tissue. Finally, the multilabel deep supervision is proposed to generate the final segmentation map. We compare with state-of-the-art medical image segmentation methods on the medical image computing and computer-assisted intervention (MICCAI) automated cardiac diagnosis challenge datasets and the MICCAI atrial segmentation challenge datasets. The mean dice scores of our method in the left atrium, right ventricle, myocardium, and left ventricle are 0.919, 0.920, 0.881, and 0.960, respectively. The analysis of evaluation indicators and segmentation results shows that our method achieves the best performance in cardiac magnetic resonance images segmentation.
- Open Access
- 21 October 2022
- Published by Association for Computing Machinery (ACM)
- p. 185-190
Color Search: An efficient region proposal generation method
Inspired by the process of searching for targets by the human eyes, our proposed target candidate region generation method focuses on the color information of the target. Current mainstream target detection algorithms often incorporate an Region Proposal Network (RPN) component for finding out the possible locations and sizes of targets, which facilitates the training and optimization of the backbone network. The RPN network uses more convolutional layers, which makes it impossible to be applied in embedded platforms that require high real-time performance and does not take full advantage of the target's a priori information. In this paper, We convert the image from RGB color model to HSV model and reduce dimensionality to form a color set. After quantizing the color set, we can obtain the Q-HSV of the image by which to find out the possible locations of the target efficiently. With this method, we can find out the possible areas of the target using only one RGB template of the target, and it is not affected by target viewpoint changes, shape changes, etc. Our approach can significantly reduce the image data to be processed by recognition and detection algorithms by more than 80%.
- 21 October 2022
- Published by Association for Computing Machinery (ACM)
- p. 185-190
Open AccessColor Search: An efficient region proposal generation method
Inspired by the process of searching for targets by the human eyes, our proposed target candidate region generation method focuses on the color information of the target. Current mainstream target detection algorithms often incorporate an Region Proposal Network (RPN) component for finding out the possible locations and sizes of targets, which facilitates the training and optimization of the backbone network. The RPN network uses more convolutional layers, which makes it impossible to be applied in embedded platforms that require high real-time performance and does not take full advantage of the target's a priori information. In this paper, We convert the image from RGB color model to HSV model and reduce dimensionality to form a color set. After quantizing the color set, we can obtain the Q-HSV of the image by which to find out the possible locations of the target efficiently. With this method, we can find out the possible areas of the target using only one RGB template of the target, and it is not affected by target viewpoint changes, shape changes, etc. Our approach can significantly reduce the image data to be processed by recognition and detection algorithms by more than 80%.
- Open Access
A Fast Adaptive Multi-Scale Kernel Correlation Filter Tracker for Rigid Object
The efficient and accurate tracking of a target in complex scenes has always been one of the challenges to tackle. At present, the most effective tracking algorithms are basically neural network models based on deep learning. Although such algorithms have high tracking accuracy, the huge number of parameters and computations in the network models makes it difficult for such algorithms to meet the real-time requirements under limited hardware conditions, such as embedded platforms with small size, low power consumption and limited computing power. Tracking algorithms based on a kernel correlation filter are well-known and widely applied because of their high performance and speed, but when the target is in a complex background, it still can not adapt to the target scale change and occlusion, which will lead to template drift. In this paper, a fast multi-scale kernel correlation filter tracker based on adaptive template updating is proposed for common rigid targets. We introduce a simple scale pyramid on the basis of Kernel Correlation Filtering (KCF), which can adapt to the change in target size while ensuring the speed of operation. We propose an adaptive template updater based on the Mean of Cumulative Maximum Response Values (MCMRV) to alleviate the problem of template drift effectively when occlusion occurs. Extensive experiments have demonstrated the effectiveness of our method on various datasets and significantly outperformed other state-of-the-art methods based on a kernel correlation filter.
Open AccessA Fast Adaptive Multi-Scale Kernel Correlation Filter Tracker for Rigid Object
The efficient and accurate tracking of a target in complex scenes has always been one of the challenges to tackle. At present, the most effective tracking algorithms are basically neural network models based on deep learning. Although such algorithms have high tracking accuracy, the huge number of parameters and computations in the network models makes it difficult for such algorithms to meet the real-time requirements under limited hardware conditions, such as embedded platforms with small size, low power consumption and limited computing power. Tracking algorithms based on a kernel correlation filter are well-known and widely applied because of their high performance and speed, but when the target is in a complex background, it still can not adapt to the target scale change and occlusion, which will lead to template drift. In this paper, a fast multi-scale kernel correlation filter tracker based on adaptive template updating is proposed for common rigid targets. We introduce a simple scale pyramid on the basis of Kernel Correlation Filtering (KCF), which can adapt to the change in target size while ensuring the speed of operation. We propose an adaptive template updater based on the Mean of Cumulative Maximum Response Values (MCMRV) to alleviate the problem of template drift effectively when occlusion occurs. Extensive experiments have demonstrated the effectiveness of our method on various datasets and significantly outperformed other state-of-the-art methods based on a kernel correlation filter.