Search

Object counting, whose aim is to estimate the number of objects from a given image, is an important and challenging computation task. Significant efforts have been devoted to addressing this problem and achieved great progress, yet counting the number of ground objects from remote sensing images is barely studied. In this paper, we are interested in counting dense objects from remote sensing images. Compared with object counting in a natural scene, this task is challenging in the following factors: large scale variation, complex cluttered background, and orientation arbitrariness. More importantly, the scarcity of data severely limits the development of research in this field. To address these issues, we first construct a large-scale object counting dataset with remote sensing images, which contains four important geographic objects: buildings, crowded ships in harbors, large-vehicles and small-vehicles in parking lots. We then benchmark the dataset by designing a novel neural network that can generate a density map of an input image. The proposed network consists of three parts namely attention module, scale pyramid module and deformable convolution module to attack the aforementioned challenging factors. Extensive experiments are performed on the proposed dataset and one crowd counting datset, which demonstrate the challenges of the proposed dataset and the superiority and effectiveness of our method compared with state-of-the-art methods., To be prepared in TGRS

Co-saliency detection aims to detect common salient objects from a group of relevant images. Some attempts have been made with the Fully Convolutional Network (FCN) framework and achieve satisfactory detection results. However, due to stacking convolution layers and pooling operation, the boundary details tend to be lost. In addition, existing models often utilize the extracted features without discrimination, leading to redundancy in representation since actually not all features are helpful to the final prediction and some even bring distraction. In this paper, we propose a co-attention module embedded FCN framework, called as Co-Attention FCN (CA-FCN). Specifically, the co-attention module is plugged into the high-level convolution layers of FCN, which can assign larger attention weights on the common salient objects and smaller ones on the background and uncommon distractors to boost final detection performance. Extensive experiments on three popular co-saliency benchmark datasets demonstrate the superiority of the proposed CA-FCN, which outperforms state-of-the-arts in most cases. Besides, the effectiveness of our new co-attention module is also validated with ablation studies.

Image dehazing aims to restore the blurry image information caused by the ambiguities of unknown scene radiance and transmission. Instead of using paired images or depth information, we propose an Unsupervised Conditional Disentangle Network (UCDN) using unpaired dataset. Our approach enforces the constraint by introducing physical-based disentanglement. Unlike other unsupervised dehazing models, our approach adapts the multi-concentration of fog and outperforms on the dataset with different concentrations. Extensive experiments on synthesized dataset demonstrate that our approach can surpass state-of-the-arts. Meanwhile, through benchmarking on our collected natural hazy dataset, our approach can generate more perceptually appealing dehazing results.

Salient object detection (SOD) is a fundamental task in computer vision, which attempts to mimic human visual systems that rapidly respond to visual stimuli and locate visually salient objects in various scenes. Perceptual studies have revealed that visual contrast is the most important factor in bottom-up visual attention process. Many of the proposed models predict saliency maps based on the computation of visual contrast between salient regions and backgrounds. In this paper, we design an end-to-end multi-scale global contrast convolutional neural network (CNN) that explicitly learns hierarchical contrast information among global and local features of an image to infer its salient object regions. In contrast to many previous CNN based saliency methods that apply super-pixel segmentation to obtain homogeneous regions and then extract their CNN features before producing saliency maps region-wise, our network is pre-processing free without any additional stages, yet it predicts accurate pixel-wise saliency maps. Extensive experiments demonstrate that the proposed network generates high quality saliency maps that are comparable or even superior to those of state-of-the-art salient object detection architectures.

Estimating accurate number of interested objects from a given image is a challenging yet important task. Significant efforts have been made to address this problem and achieve great progress, yet counting number of ground objects from remote sensing images is barely studied. In this paper, we are interested in counting dense objects from remote sensing images. Compared with object counting in natural scene, this task is challenging in following factors: large scale variation, complex cluttered background and orientation arbitrariness. More importantly, the scarcity of data severely limits the development of research in this field. To address these issues, we first construct a large-scale object counting dataset based on remote sensing images, which contains four kinds of objects: buildings, crowded ships in harbor, large-vehicles and small-vehicles in parking lot. We then benchmark the dataset by designing a novel neural network which can generate density map of an input image. The proposed network consists of three parts namely convolution block attention module (CBAM), scale pyramid module (SPM) and deformable convolution module (DCM). Experiments on the proposed dataset and comparisons with state of the art methods demonstrate the challenging of the proposed dataset, and superiority and effectiveness of our method.

Fabric defect detection plays an important role in the quality control of textile products. Most existing defect detection techniques adopted traditional pattern recognition methods, which were lacking adaptability and presented the undesirable detection accuracy. In this paper, a fabric defect detection algorithm based on multi-channel feature matrixes extraction and joint low-rank decomposition was proposed by simulating biological visual perception mechanism. Based on the fact that the second-order gradient information is more suitable for characterizing the fabric texture, we developed a novel second-order multi-channel feature extraction method by modeling the response and distribution properties of the P-type ganglion cells in the primate retina. Upon devising a powerful descriptor, a joint low-rank decomposition method is utilized to model biological visual saliency, and decomposes the fabric images into backgrounds and salient defect objects. Experimental results demonstrate that our proposed algorithm has good self-adaptability and detection performance for plain and twill fabrics or complex patterned fabrics, and is superior to the state-of-the-art methods.

Fabric defect detection plays a key role in the quality control of textiles. Existing fabric defect detection methods adopt traditional pattern recognition methods; however, these methods lack adaptability and present poor detection performance. Because biological vision system has the ability to quickly locate salient objects, we propose a novel fabric defect detection algorithm based on biological vision modeling by simulating the mechanism of biological visual perception. First, a distinct, efficient, and robust feature descriptor from the biological modeling of P ganglion cells, which was proposed in our previous work, is adopted to improve the representation of fabric images with complex textures. To account for the low-rank and sparsity characteristics of biological vision, the low-rank representation (LRR) technique is adopted to model biological visual saliency, and it can decompose the fabric image into backgrounds and salient defect objects. Meanwhile, dictionary learning and Laplacian regularization are integrated into the LRR model as follows: 1) dictionary learning is used to denoise the saliency map; and 2) Laplacian regularization enlarges the gaps between defective regions and the background. Finally, the linearized accelerated direction method with adaptive penalty is adopted to solve the proposed model. The experimental results emphasize that the proposed algorithm has good detection performance for plain or twill fabrics with simple textures as well as for patterned fabrics with complex textures. Moreover, the proposed method is superior to the state-of-the-art methods in terms of its adaptability and detection efficiency.

Purpose The purpose of this paper is to find an efficient fabric defect detection algorithm by means of exploring the sparsity characteristics of main local binary pattern (MLBP) extracted from the original fabric texture. Design/methodology/approach In the proposed algorithm, original LBP features are extracted from the fabric texture to be detected, and MLBP are selected by occurrence probability. Second, a dictionary is established with MLBP atoms which can sparsely represent all the LBP. Then, the value of the gray-scale difference between gray level of neighborhood pixels and the central pixel, and the mean of the difference which has the same MLBP feature are calculated. And then, the defect-contained image is reconstructed as normal texture image. Finally, the residual is calculated between reconstructed and original images, and a simple threshold segmentation method can divide the residual image, and the defective region is detected. Findings The experiment result shows that the fabric texture can be more efficiently reconstructed, and the proposed method achieves better defect detection performance. Moreover, it offers empirical insights about how to exploit the sparsity of one certain feature, e.g. LBP. Research limitations/implications Because of the selected research approach, the results may lack generalizability in chambray. Therefore, researchers are encouraged to test the proposed propositions further. Originality/value In this paper, a novel fabric defect detection method which extracts the sparsity of MLBP features is proposed.

Purpose – Fabric defect detection plays an important role in textile quality control. The purpose of this paper is to propose a fabric defect detection algorithm using learned dictionary-based visual saliency. Design/methodology/approach – First, the test fabric image is splitted into image blocks, and the learned dictionary with normal samples and defective sample is constructed by selecting the image block local binary pattern features with highest or lowest similarity comparing with the average feature vector; second, the first L largest correlation coefficients between each test image block and the dictionary are calculated, and other correlation coefficients are set to zeros; third, the sum of the non-zeros coefficients corresponding to defective samples is used to generate saliency map; finally, an improve valley-emphasis method can efficiently segment the defect region. Findings – Experimental results demonstrate that the generated saliency map by the proposed method can efficiently outstand defect region comparing with the state-of-the-art, and segment results can precisely localize defect region. Originality/value – In this paper, a novel fabric defect detection scheme is proposed via learned dictionary-based visual saliency.

Fabric defect detection plays a curial step in the quality control of textiles. Existing fabric defect detection methods are lack of adaptability and have a poor detection performance. A novel fabric defect detection method based on Gabor filter and tensor low-rank recovery was proposed in this paper. Defect-free fabric images have the specified direction, while defects damage their regularity of direction. Therefore, the direction feature is curial for fabric defect detection. For different kinds of fabric image, the direction information is also distinct. In order to characterize the direction information for all kinds of fabric image, we adopted a bank of Gabor directional filters to extract directional information, and generated the directional Gabor filtered maps. Thereafter, an efficient TRPCA model is proposed to decompose the feature tensor which is generated by stacking the feature vector of all the feature maps into a low-rank tensor and a sparse tensor by the alternating direction method of multipliers according to the tensor recovery (ADMM-TR) techniques. Finally, the saliency map generated by the sparse tensor part is segmented via the improved adaptive thresholding algorithm to locate the defective regions. Experimental results demonstrate that our algorithm is superior to the state-of-the-art.

Fabric defect detection plays an important role in the quality control of fabric products. In order to effectively detect defects for fabric images with numerous kinds of defects and complex texture, a novel fabric defect detection algorithm based on multi-channel feature extraction and joint low-rank decomposition is proposed. First, at the feature extraction stage, a multi-channel robust feature (Multi-channel Distinctive Efficient Robust Feature, McDerf) is extracted by simulating the biological visual perception mechanism for multiple gradient orientation maps. Second, joint low-rank decomposition algorithm is adopted to decompose the feature matrix into a low rank matrix and a sparse matrix. Finally, for the purpose of localizing the defect region, the threshold segmentation algorithm is utilized to segment the saliency map generated by sparse matrix. Comparing with the existing fabric defect detection algorithms, the experimental results show that the proposed algorithm has better adaptability and detection efficiency for the plain and patterned fabric images.

In order to accurately detect the fabric defects in production process, an effective fabric detection algorithm based on Gabor filter and low-rank decomposition is proposed. Firstly, the Gabor filter features with multi-scale and multiple directions are extracted from the fabric image, then the extracted Gabor feature maps are divided into the blocks with size 16×16 by uniform sampling; secondly, we calculate the average feature vector for each block, and stack the feature vectors of all blocks into a feature matrix; thirdly, an efficient low rank decomposition model is built for feature matrix, and is divided into a low-rank matrix and a sparse matrix by the accelerated proximal gradient approach (APG). Finally, the saliency map generated by sparse matrix is segmented by the improved optimal threshold algorithm, to locate the defect regions. Experiment results show that low-rank decomposition can effectively detect fabric defect, and outperforms the state-of-the-art methods.