In the existing one-factor cancelable biometric template protection scheme, the hashing function used in the transformation of biometrics can’t preserve the original biometric features, which leads to low recognition rate. To make full use of biometric features by replication and extension, but too long feature vectors can cause low computational efficiency. Therefore, a one-factor cancelable fingerprint template protection based on feature enhanced hashing is proposed. Firstly, the extended binary biometric vectors are combined by sliding and extracting window, then converted into decimal system, in order to make full use of biometric features and increase non-invertibility. Secondly, the permutation factor is calculated by the feature enhanced hashing function and the random sequence is reordered, it can embed the information of the original biometric features into the random sequence better. Finally, a cancelable template is generated by reducing the same length of the first and last of reordered random sequence, in this way, some elements can be deleted to improve the computational efficiency and non-invertibility. The experimental results show that the recognition rate of the algorithm is improved on FVC2002 and FVC2004 fingerprint databases, which meets the design standards of cancelable biometric recognition and can defend against security attacks.
Methods based on stacked hourglass networks (SHNs) have achieved great progress in face alignment tasks. However, most of these algorithms have met with limited success in modeling correlations among features. Visual attention mechanisms have shown promise in terms of effectively understanding scenes in various computer vision tasks. In this paper, an attention-guided coarse-to-fine network (AGCFN) based on an attention mechanism is proposed for robust face alignment. Thus, the network is guided to emphasize key information while suppressing less important information. Meanwhile, the fusion of features from different levels is adopted to improve the information flow through the proposed network. Additionally, conditional random fields (CRFs) are introduced to model the spatial interactions between landmarks in the prediction maps. Experimental results obtained on the 300-W dataset, the 300-W private test set, and the WFLW dataset demonstrate the superiority of the proposed method in terms of accuracy and robustness.
Traditional methods of extracting finger vein texture changes and orientation features are susceptible to illumination, translation, noise, and rotation, and the process has difficulty directly extracting structural features through the original image. In this paper, the histogram of competitive Gabor directional binary statistics (HCGDBS) is proposed to extract discriminative structural features. First, the index of the largest filter value is obtained based on the multidirectional Gabor filter as the dominant direction, thereby obtaining the rotation-invariance feature. Second, according to the filter response size of each pixel in different directions, the order difference relationship between the adjacent three directions is compared, and a highly discriminative competitive Gabor direction binary pattern (CGDBP) is constructed. Finally, the CGDBP features are extracted in blocks, and the HCGDBS is constructed to overcome image translation. Experimental results show that it improves the recognition performance and overcomes illumination, translation, noise, and rotation.
Aiming at the problems that blurred edge structure, loss of texture details, distortion and slow running speed in medical image fusion, a novel medical image fusion method based on adaptive weighted guided image filtering is proposed in this paper. First of all, the weight factor of the weighted guided image filtering is optimized by using the gradient operator, and the gradient operator weighted guided image filtering with better edge-preserving ability is obtained. The filter is used to decompose the MRI image into smooth layer and detail layer. Secondly, a soft threshold fusion rule is designed to fuse the smooth layer and the PET/SPECT image to obtain the fusion sub-image, which can retain the contour structure and color information of the original image. Finally, the selective enhancement of the detail layer of MRI image can improve the texture information, and then combine the enhanced detail layer with the fusion sub-image to obtain the final fusion image. The experimental results show that the subjective visual effect of the fused image contains the basic spatial structure, rich texture details and color information. The overall performance of the objective evaluation index is also better than the algorithm compared.
This paper proposes a new method to deal with visual information expression of image formation and systematization based on visual information representation theory, and analyzes the characteristics of multi-source remote sensing image from the perspective of remote sensing imaging mechanism, and expatiates some pivotal rules regarding visual information during image capturing, description and reconstruction. In the process of formulating SIFT description, this paper makes a detailed research on how to calculate the lower matching pints and marginal points, adjust the threshold of the feature matching parameters and increase the matching points numbers automatically, based on the number of the exiting matching points and their distribution conditions. In this experiment, the number of feature points increases with the decrease of the threshold of low contrast points, and edge response points, which shows the similar changes in the Law of Inverse; While in the process of automatic matching, the number of feature points increases with the increase of radio value of the farthest distance of the feature points to the nearest distance, showing almost directly proportional to the law. In general, as the number of matching points increase, the accuracy and the stability of the matching would decrease. This paper proposes a threshold weight of the adaptive algorithm to improve the accuracy and robustness of the matching points and solves the problems described above. Therefore, the multi-source remote sensing images are generally divided into the images with same resolution and those with different resolutions. When the reference image and the uncorrected image have the same resolution, the connection lines of the matching points will have the same distance and slope. By contract, when the resolution of the reference image and that of the uncorrected images are different, their connection lines of matching points will intersect. This paper, studying this geometric constraint conditions, suggests a fast mismatching points' rejected method based on rough fuzzy C-Means cluster theory. This paper then discusses the precise matching of residual matching points using Least Square Method. Numerous experiments are conducted for both aerial and satellite imageries under various conditions such as geometric distortion, illumination variation and different resolutions. Results of this study show that the proposed matching approach performs well, and the matching accuracy is stable and reliable.
Recognition of facial expression is a challenge when it comes to computer vision. The primary reasons are class imbalance due to data collection and uncertainty due to inherent noise such as fuzzy facial expressions and inconsistent labels. However, current research has focused either on the problem of class imbalance or on the problem of uncertainty, ignoring the intersection of how to address these two problems. Therefore, in this paper, we propose a framework based on Resnet and Attention to solve the above problems. We design weight for each class. Through the penalty mechanism, our model will pay more attention to the learning of small samples during training, and the resulting decrease in model accuracy can be improved by a Convolutional Block Attention Module (CBAM). Meanwhile, our backbone network will also learn an uncertain feature for each sample. By mixing uncertain features between samples, the model can better learn those features that can be used for classification, thus suppressing uncertainty. Experiments show that our method surpasses most basic methods in terms of accuracy on facial expression data sets (e.g., AffectNet, RAF-DB), and it also solves the problem of class imbalance well.
Synthetic aperture radar (SAR) images reveal severe geometric distortions especially in the mountain area, such as layover, which is caused by imaging characteristics of SAR itself and terrain undulations. The layover phenomenon greatly limits the application of SAR images. This article proposes a layover compensation method for regional spaceborne SAR imagery without ground control points (GCPs), which is mainly improved from two aspects. First, a method based on rational function model (RFM) to determine the layover range is proposed. Second, based on geometric calibration and block adjustment, the processing flow is optimized to generate digital orthophoto map (DOM) which greatly eliminated the influence of layover. The proposed method was applied to Chinese Gaofen-3 (GF-3) SAR regional images, including the ascending and descending track stacks. The result showed that 84.5% of the layover pixels on the regional DOM were compensated, which verified the effectiveness and feasibility of the method.
'/%3e%3cuse xlink:href='%23a' transform='rotate(23.036 .093 25.536)'/%3e%3cuse xlink:href='%23a' transform='rotate(45.87 1.273 16.18)'/%3e%3cuse xlink:href='%23a' transform='rotate(69.945 .996 12.078)'/%3e%3cuse xlink:href='%23a' transform='rotate(20.66 -19.689 31.932)'/%3e%3c/svg%3e)