This page has only limited features, please log in for full access.
In the past few years, multitask learning (MTL) has been widely used in a single model to solve the problems of multiple businesses. MTL enables each task to achieve high performance and greatly reduces computational resource overhead. In this work, we designed a collaborative network that simultaneously solves the super-resolution semantic segmentation and super-resolution image reconstruction. This algorithm can obtain high-resolution semantic segmentation and super-resolution reconstruction results by taking relatively low-resolution images as input when high-resolution data are inconvenient or computing resources are limited. The framework consists of three parts: the semantic segmentation branch (SSB), the super-resolution branch (SRB), and the structural affinity block (SAB). Specifically, the SSB, SRB, and SAB are responsible for completing super-resolution semantic segmentation, image super-resolution reconstruction, and associated features, respectively. Our proposed method is simple and efficient, and it can replace the different branches with most of the state-of-the-art models. The International Society for Photogrammetry and Remote Sensing (ISPRS) segmentation benchmarks were used to evaluate our models. In particular, super-resolution semantic segmentation on the Potsdam dataset reduced Intersection over Union (IoU) by only 1.8% when the resolution of the input image was reduced by a factor of two. The experimental results showed that our framework can obtain more accurate semantic segmentation and super-resolution reconstruction results than the single model.
Qian Zhang; Guang Yang; Guixu Zhang. Collaborative Network for Super-Resolution and Semantic Segmentation of Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing 2021, PP, 1 -12.
AMA StyleQian Zhang, Guang Yang, Guixu Zhang. Collaborative Network for Super-Resolution and Semantic Segmentation of Remote Sensing Images. IEEE Transactions on Geoscience and Remote Sensing. 2021; PP (99):1-12.
Chicago/Turabian StyleQian Zhang; Guang Yang; Guixu Zhang. 2021. "Collaborative Network for Super-Resolution and Semantic Segmentation of Remote Sensing Images." IEEE Transactions on Geoscience and Remote Sensing PP, no. 99: 1-12.
Identifying interactions between compound and protein is a substantial part of the drug discovery process. Accurate prediction of interaction relationships can greatly reduce the time of drug development. The uniqueness of our method lies in three aspects:1) it represents a compound with a distance matrix. A distance matrix can capture the structural information, compared with the SMILES string. On the other hand, a distance matrix does not require complex data preprocessing for the molecular structure as the molecular graph representation, and is easier to obtain; 2) it uses SPP(Spatial pyramid pooling)-net to extract compound features, which has been successfully applied in image classification; and 3) it extracts protein features through the natural language processing method (doc2vec) to obtain sequence semantic information. We evaluated our method on three benchmark datasets-human, C.elegans, and DUDE and the experimental results demonstrate that our proposed model presents competitive performance against state-of-the-art predictors. We also carried out drug-drug interaction (DDI) experiments to verify the strong potential of distance matrix as molecular characteristics. The source code and datasets are available at https://github.com/lxlsu/SPP_CPI.
Ying Qian; Xuelian Li; Qian Zhang; Jiongmin Zhang. SPP-CPI: Predicting Compound-Protein Interactions Based on Neural Networks. IEEE/ACM Transactions on Computational Biology and Bioinformatics 2021, PP, 1 -1.
AMA StyleYing Qian, Xuelian Li, Qian Zhang, Jiongmin Zhang. SPP-CPI: Predicting Compound-Protein Interactions Based on Neural Networks. IEEE/ACM Transactions on Computational Biology and Bioinformatics. 2021; PP (99):1-1.
Chicago/Turabian StyleYing Qian; Xuelian Li; Qian Zhang; Jiongmin Zhang. 2021. "SPP-CPI: Predicting Compound-Protein Interactions Based on Neural Networks." IEEE/ACM Transactions on Computational Biology and Bioinformatics PP, no. 99: 1-1.
Semantic segmentation is a fundamental task in remote sensing image processing. It provides pixel-level classification, which is important for many applications, such as building extraction and land use mapping. The development of convolutional neural network has considerably improved the performance of semantic segmentation. Most semantic segmentation networks are the encoder-decoder structure. Bilinear interpolation is an ordinary upsampling method in the decoder, but bilinear interpolation only considers its own features and inserts three times its own features. This over-simple and data-independent bilinear upsampling may lead to suboptimal results. In this work, we propose an upsampling method based on local relations to replace bilinear interpolation. Upsampling is performed by correlating the local relationship of feature maps of adjacent stages, which can better integrate local and global information. We also design a fusion module based on local similarity. Our proposed method with ResNet101 as the backbone of the segmentation network can improve the average F₁ score and overall accuracy of the Vaihingen data set by 2.69% and 1.31%, respectively. Our proposed method also has fewer parameters and less inference time.
Baokai Lin; Guang Yang; Qian Zhang; Guixu Zhang. Semantic Segmentation Network Using Local Relationship Upsampling for Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters 2021, PP, 1 -5.
AMA StyleBaokai Lin, Guang Yang, Qian Zhang, Guixu Zhang. Semantic Segmentation Network Using Local Relationship Upsampling for Remote Sensing Images. IEEE Geoscience and Remote Sensing Letters. 2021; PP (99):1-5.
Chicago/Turabian StyleBaokai Lin; Guang Yang; Qian Zhang; Guixu Zhang. 2021. "Semantic Segmentation Network Using Local Relationship Upsampling for Remote Sensing Images." IEEE Geoscience and Remote Sensing Letters PP, no. 99: 1-5.
Deep learning methods have been used to extract buildings from remote sensing images and have achieved state-of-the-art performance. Most previous work has emphasized the multi-scale fusion of features or the enhancement of more receptive fields to achieve global features rather than focusing on low-level details such as the edges. In this work, we propose a novel end-to-end edge-aware network, the EANet, and an edge-aware loss for getting accurate buildings from aerial images. Specifically, the architecture is composed of image segmentation networks and edge perception networks that, respectively, take charge of building prediction and edge investigation. The International Society for Photogrammetry and Remote Sensing (ISPRS) Potsdam segmentation benchmark and the Wuhan University (WHU) building benchmark were used to evaluate our approach, which, respectively, was found to achieve 90.19% and 93.33% intersection-over-union and top performance without using additional datasets, data augmentation, and post-processing. The EANet is effective in extracting buildings from aerial images, which shows that the quality of image segmentation can be improved by focusing on edge details.
Guang Yang; Qian Zhang; Guixu Zhang. EANet: Edge-Aware Network for the Extraction of Buildings from Aerial Images. Remote Sensing 2020, 12, 2161 .
AMA StyleGuang Yang, Qian Zhang, Guixu Zhang. EANet: Edge-Aware Network for the Extraction of Buildings from Aerial Images. Remote Sensing. 2020; 12 (13):2161.
Chicago/Turabian StyleGuang Yang; Qian Zhang; Guixu Zhang. 2020. "EANet: Edge-Aware Network for the Extraction of Buildings from Aerial Images." Remote Sensing 12, no. 13: 2161.
Deep learning now plays an important role in solving complex problems in computer vision fields. The highly challenging high-resolution remote sensing image scene classification problem can also be solved using deep learning methods. The most commonly used method of deep learning is the convolutional neural network model. In this letter, based on deep learning, a combined model named Inception-long short-term memory (LSTM) is proposed. First, we combine the deep learning feature extracted from the pretrained Inception-V3 model with a hand-crafted feature: the GIST feature. The different features are then combined and input into the batch normalization (BN) layer. Second, the BN layer plays the role of the bridge to combine the InceptionV3 model with the LSTM model, which features a softmax classifier. The LSTM model is used to analyze the features and classify the different high-resolution remote sensing scene images. The proposed model, as a whole, can be uniformly trained. Three different datasets--the NWPU-RESISC45 dataset, the UC Merced dataset, and the SIRI-WHU dataset--were used to verify the effectiveness of the proposed model. The results show that the proposed Inception-LSTM model shows an outstanding performance in the scene classification task.
Yunya Dong; Qian Zhang. A Combined Deep Learning Model for the Scene Classification of High-Resolution Remote Sensing Image. IEEE Geoscience and Remote Sensing Letters 2019, 16, 1540 -1544.
AMA StyleYunya Dong, Qian Zhang. A Combined Deep Learning Model for the Scene Classification of High-Resolution Remote Sensing Image. IEEE Geoscience and Remote Sensing Letters. 2019; 16 (10):1540-1544.
Chicago/Turabian StyleYunya Dong; Qian Zhang. 2019. "A Combined Deep Learning Model for the Scene Classification of High-Resolution Remote Sensing Image." IEEE Geoscience and Remote Sensing Letters 16, no. 10: 1540-1544.
Halogen bonds (XBs) are attracting increasing attention in biological systems. Protein Data Bank (PDB) archives experimentally determined XBs in biological macromolecules. However, no software for structure refinement in X-ray crystallography takes into account XBs, which might result in the weakening or even vanishing of experimentally determined XBs in PDB. In our previous study, we showed that side-chain XBs forming with protein side chains are underestimated in PDB on the basis of the phenomenon that the proportion of side-chain XBs to overall XBs decreases as structural resolution becomes lower and lower. However, whether the dominant backbone XBs forming with protein backbone are overlooked is still a mystery. Here, with the help of the ratio (RF) of the observed XBs’ frequency of occurrence to their frequency expected at random, we demonstrated that backbone XBs are largely overlooked in PDB, too. Furthermore, three cases were discovered possessing backbone XBs in high resolution structures while losing the XBs in low resolution structures. In the last two cases, even at 1.80 Å resolution, the backbone XBs were lost, manifesting the urgent need to consider XBs in the refinement process during X-ray crystallography study.
Qian Zhang; Zhijian Xu; Jiye Shi; Weiliang Zhu. Underestimated Halogen Bonds Forming with Protein Backbone in Protein Data Bank. Journal of Chemical Information and Modeling 2017, 57, 1529 -1534.
AMA StyleQian Zhang, Zhijian Xu, Jiye Shi, Weiliang Zhu. Underestimated Halogen Bonds Forming with Protein Backbone in Protein Data Bank. Journal of Chemical Information and Modeling. 2017; 57 (7):1529-1534.
Chicago/Turabian StyleQian Zhang; Zhijian Xu; Jiye Shi; Weiliang Zhu. 2017. "Underestimated Halogen Bonds Forming with Protein Backbone in Protein Data Bank." Journal of Chemical Information and Modeling 57, no. 7: 1529-1534.
Urban areas are a complex combination of various land-cover types, and show a variety of land-use structures and spatial layouts. Furthermore, the spectral similarity between built-up areas and bare land is a great challenge when using high spatial resolution remote sensing images to map urban areas, especially for images obtained in dry and cold seasons or high-latitude regions. In this study, a new procedure for urban area extraction is presented based on the high-level, regional, and line segment features of high spatial resolution satellite data. The urban morphology is also analyzed. Firstly, the primitive features—the morphological building index (MBI), the normalized difference vegetation index (NDVI), and line segments—are extracted from the original images. Chessboard segmentation is then used to segment the image into the same-size objects. In each object, advanced features are then extracted based on the MBI, the NDVI, and the line segments. Subsequently, object-oriented classification is implemented using the above features to distinguish urban areas from non-urban areas. In general, the boundaries of urban and non-urban areas are not very clear, and each urban area has its own spatial structure characteristic. Hence, in this study, an analysis of the urban morphology is carried out to obtain a clear regional structure, showing the main city, the surrounding new development zones, etc. The experimental results obtained with six WorldView-2 and Gaofen-2 images obtained from different regions and seasons demonstrate that the proposed method outperforms the current state-of-the-art methods.
Qian Zhang; Xin Huang; Guixu Zhang. Urban Area Extraction by Regional and Line Segment Feature Fusion and Urban Morphology Analysis. Remote Sensing 2017, 9, 663 .
AMA StyleQian Zhang, Xin Huang, Guixu Zhang. Urban Area Extraction by Regional and Line Segment Feature Fusion and Urban Morphology Analysis. Remote Sensing. 2017; 9 (7):663.
Chicago/Turabian StyleQian Zhang; Xin Huang; Guixu Zhang. 2017. "Urban Area Extraction by Regional and Line Segment Feature Fusion and Urban Morphology Analysis." Remote Sensing 9, no. 7: 663.
Halogen bonds (XBs) have been attracting increasing attention in biological systems, especially in drug discovery and design, for their advantages of both improving drug-target binding affinity and tuning ADME/T properties. After a comprehensive literature survey in drug discovery and design, we found that most of the studies on XBs between ligands and proteins have focused on the protein backbone. Meanwhile, we also noticed that the proportion of side-chain XBs to overall XBs decreases as structural resolution becomes lower and lower. We postulated that protein side chains are more flexible in comparison with backbone structures, leading to more unclear electron density and lower resolution of the side chains. As the classic force field used to refine protein structures from diffraction data cannot handle XBs correctly, some of the interactions are lost during the refinement. On the contrary, there is no change in the corresponding ratio of hydrogen bonds (HBs) during structural resolution because HBs can be handled well with the classic force field. Further analysis revealed that Thr and Gln account for a large part of the decreasing XB trend, which could be partly attributed to the misidentified N, C, or O atoms. In addition, the lost XBs might be recovered after the atoms are reassigned, e.g., by flipping Thr side chains. In summary, formation of XBs with protein side chains is underestimated, and more attention should be paid to the potential formation of XBs between organohalogens and protein side chains during X-ray crystallography studies.
Qian Zhang; Zhijian Xu; Weiliang Zhu. The Underestimated Halogen Bonds Forming with Protein Side Chains in Drug Discovery and Design. Journal of Chemical Information and Modeling 2016, 57, 22 -26.
AMA StyleQian Zhang, Zhijian Xu, Weiliang Zhu. The Underestimated Halogen Bonds Forming with Protein Side Chains in Drug Discovery and Design. Journal of Chemical Information and Modeling. 2016; 57 (1):22-26.
Chicago/Turabian StyleQian Zhang; Zhijian Xu; Weiliang Zhu. 2016. "The Underestimated Halogen Bonds Forming with Protein Side Chains in Drug Discovery and Design." Journal of Chemical Information and Modeling 57, no. 1: 22-26.
This letter proposes an efficient framework for building detection from coarse to fine using morphological technique for high-resolution optical satellite imagery over urban areas. First, the preliminary result of building regions is obtained by the recently developed morphological building index (MBI) method, which is able to detect potential building structures. However, the raw results derived from the MBI can be subject to a number of false alarms, which are caused by bright soil, roads, and open areas. In this letter, we propose to use morphological spatial pattern analysis as a postprocessing to further optimize the MBI result and remove the commission errors. The original MBI result is then separated into seven mutually exclusive categories-core, islet, loop, bridge, perforation, edge, and branch-by applying a series of morphological transformations such as erosions, geodesic dilation, reconstruction by dilation, anchored skeletonization, etc. The objects corresponding to the generic categories are then analyzed, and the categories corresponding to building parts are maintained, while the others are abandoned. After this postprocessing, the small noisy patches and narrow roads, which were wrongly extracted by the MBI, can be removed. In addition, the shape of the buildings can also be regularized by removing the branches, and the holes contained in the building objects can be identified and filled. Extensive experiments performed on GeoEye-1 and WorldView-2 images confirm the effectiveness and robustness of the proposed morphological building detection framework.
Qian Zhang; Xin Huang; Guixu Zhang. A Morphological Building Detection Framework for High-Resolution Optical Imagery Over Urban Areas. IEEE Geoscience and Remote Sensing Letters 2016, 13, 1388 -1392.
AMA StyleQian Zhang, Xin Huang, Guixu Zhang. A Morphological Building Detection Framework for High-Resolution Optical Imagery Over Urban Areas. IEEE Geoscience and Remote Sensing Letters. 2016; 13 (9):1388-1392.
Chicago/Turabian StyleQian Zhang; Xin Huang; Guixu Zhang. 2016. "A Morphological Building Detection Framework for High-Resolution Optical Imagery Over Urban Areas." IEEE Geoscience and Remote Sensing Letters 13, no. 9: 1388-1392.
Pan-sharpening is a technique that generates a high spatial resolution multi-spectral image making use of both the spectral information contained in a low spatial resolution multi-spectral image and the spatial information contained in a high spatial resolution panchromatic image. The pan-sharpening method usually contains some parameters. They are usually problem dependent and need to be set properly. In this article, we propose a variational method for pan-sharpening and use an evolutionary algorithm (EA) to choose the optimal parameters automatically. In our method, two quality measurements are combined to form an optimization objective function of the EA, and the parameters are encoded as an individual vector in the EA. The optimal parameters are generated by optimizing the objective function of the EA. The new method is compared with some other variational methods using QuickBird data. We also applied the selected parameters to different images to discuss the applicable scope. The experimental results show that our method can generate a high-quality fused image, and the same parameters’ values can be used for similar images.
Yang Xiao; Faming Fang; Qian Zhang; Aimin Zhou; Guixu Zhang. Parameter selection for variational pan-sharpening by using evolutionary algorithm. Remote Sensing Letters 2015, 6, 458 -467.
AMA StyleYang Xiao, Faming Fang, Qian Zhang, Aimin Zhou, Guixu Zhang. Parameter selection for variational pan-sharpening by using evolutionary algorithm. Remote Sensing Letters. 2015; 6 (6):458-467.
Chicago/Turabian StyleYang Xiao; Faming Fang; Qian Zhang; Aimin Zhou; Guixu Zhang. 2015. "Parameter selection for variational pan-sharpening by using evolutionary algorithm." Remote Sensing Letters 6, no. 6: 458-467.
Undersegmentation or oversegmentation is a challenge faced in image segmentation methods, and it is extreme important to determine the optimal number of regions (clusters) of an image in real-world applications. In this study, we introduce an adaptive strategy to do so. The basic idea is to firstly oversegment an image by using the Mean-shift (MS) method, and then segment the obtained oversegmented results by using an evolutionary algorithm. In the second stage, a feature is extracted for each region obtained by the MS method, and a new fitness function is designed to determine the optimal number of clusters. The adaptive approach is applied to a variety of images, and the experimental results show that our method is both efficient and effective for image segmentation.
Cong Liu; Aimin Zhou; Qian Zhang; Guixu Zhang. Adaptive image segmentation by using mean‐shift and evolutionary optimisation. IET Image Processing 2014, 8, 327 -333.
AMA StyleCong Liu, Aimin Zhou, Qian Zhang, Guixu Zhang. Adaptive image segmentation by using mean‐shift and evolutionary optimisation. IET Image Processing. 2014; 8 (6):327-333.
Chicago/Turabian StyleCong Liu; Aimin Zhou; Qian Zhang; Guixu Zhang. 2014. "Adaptive image segmentation by using mean‐shift and evolutionary optimisation." IET Image Processing 8, no. 6: 327-333.
An energy-driven total variation (TV) formulation is proposed for the segmentation of high spatial resolution remote-sensing imagery. The TV model is an effective tool for image processing operations such as restoration, enhancement, reconstruction, and diffusion. Due to the relationship between the TV model and the segmentation problem, in this letter, a TV-based approach is investigated for segmentation of high-spatial-resolution remote-sensing imagery. Subsequently, an object-based classification method, i.e., majority voting, is used to classify the segmented results. In experiments, the proposed TV-based method is compared with the widely used fractal net evolution approach and the clustering segmentation methods such as the expectation–maximization and $k$-means. The performances of the segmentation and the classification are evaluated based on both thematic and geometric indices.
Qian Zhang; Xin Huang; Liangpei Zhang. An Energy-Driven Total Variation Model for Segmentation and Classification of High Spatial Resolution Remote-Sensing Imagery. IEEE Geoscience and Remote Sensing Letters 2012, 10, 125 -129.
AMA StyleQian Zhang, Xin Huang, Liangpei Zhang. An Energy-Driven Total Variation Model for Segmentation and Classification of High Spatial Resolution Remote-Sensing Imagery. IEEE Geoscience and Remote Sensing Letters. 2012; 10 (1):125-129.
Chicago/Turabian StyleQian Zhang; Xin Huang; Liangpei Zhang. 2012. "An Energy-Driven Total Variation Model for Segmentation and Classification of High Spatial Resolution Remote-Sensing Imagery." IEEE Geoscience and Remote Sensing Letters 10, no. 1: 125-129.
In this article, we propose a Distance-Weighted Markov Random Field (DwMRF) for classification of high–spatial resolution imagery. The proposed DwMRF integrates the spectral and spatial information of the image, and better coordinates the interaction between neighbours and the central pixels than the conventional Equal-weighted MRF (EwMRF). In addition, we propose a Serial Iterated Conditional Mode (SICM) method for the solution of the Markov Random Field (MRF) model. Experiments are conducted on three data sets: HYDICE data of the Mall in Washington, DC, HYMAP data of Purdue University and QuickBird data of Beijing. We compare the proposed DwMRF approach with other methods: the EwMRF, Maximum Likelihood Classification (MLC) and a multiresolution segmentation (Fractal Net Evolution Approach (FNEA)) method. Experiments show the DwMRF is robust and outperforms the other methods; furthermore, the proposed SICM method converges more rapidly than conventional Iterated Conditional Mode (ICM) and provides classification results comparable with the conventional ICM method.
Qian Zhang; Liangpei Zhang; Xin Huang. Classification of high-spatial resolution imagery based on distance-weighted Markov random field with an improved iterated conditional mode method. International Journal of Remote Sensing 2011, 32, 9843 -9868.
AMA StyleQian Zhang, Liangpei Zhang, Xin Huang. Classification of high-spatial resolution imagery based on distance-weighted Markov random field with an improved iterated conditional mode method. International Journal of Remote Sensing. 2011; 32 (24):9843-9868.
Chicago/Turabian StyleQian Zhang; Liangpei Zhang; Xin Huang. 2011. "Classification of high-spatial resolution imagery based on distance-weighted Markov random field with an improved iterated conditional mode method." International Journal of Remote Sensing 32, no. 24: 9843-9868.