查看更多>>摘要：In this paper, we propose a novel semi-supervised active salient object detection (SOD) method that actively acquires a small subset of the most discriminative and representative samples for labeling. Two main contributions have been made to prevent the method from being overwhelmed by labeling similar distributed samples. First, we design a saliency encoder-decoder with adversarial discriminator to generate a confidence map, representing the network uncertainty on the current prediction. Then, we select the least confident (discriminative) samples from the unlabeled pool to form the "candidate labeled pool". Second, we train a Variational Auto-Encoder (VAE) to select and add the most representative data from the "candidate labeled pool" into the labeled pool by comparing their corresponding features in the latent space. Within our framework, these two networks are optimized conditioned on the states of each other progressively. Experimental results on six benchmarking SOD datasets demonstrate that our annotation efficient learning based salient object detection method, reaching to 14% labeling budget, can be on par with the state-of-the-art fully-supervised deep SOD models. The source code is publicly available via our project page: https://github.com/JingZhang617/Semi- sup- active-selfsup-Learning . (c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：This study proposes a contrastive convolutional auto-encoder (contrastive CAE), a combined architecture of an auto-encoder and contrastive loss, to identify individuals with suspected COVID-19 infection using heart-rate data from participants with multiple sclerosis (MS) in the ongoing RADAR-CNS mHealth research project. Heart-rate data was remotely collected using a Fitbit wristband. COVID-19 infection was either confirmed through a positive swab test, or inferred through a self-reported set of recognised symptoms of the virus. The contrastive CAE outperforms a conventional convolutional neural network (CNN), a long short-term memory (LSTM) model, and a convolutional auto-encoder without contrastive loss (CAE). On a test set of 19 participants with MS with reported symptoms of COVID-19, each one paired with a participant with MS with no COVID-19 symptoms, the contrastive CAE achieves an unweighted average recall of 95 . 3% , a sensitivity of 100% and a specificity of 90 . 6% , an area under the receiver operating characteristic curve (AUC-ROC) of 0.944, indicating a maximum successful detection of symptoms in the given heart rate measurement period, whilst at the same time keeping a low false alarm rate. (c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：A major challenge in scene understanding is the handling of spatial relations between objects or object parts. Several descriptors dedicated to this task already exist, such as the force histogram which is a typical example of relative position descriptor. By computing the interaction between two objects for a given force in all the directions, it gives a good overview of the configuration, and it has useful properties that can make it invariant to the 2D viewpoint. Considering that using complementary forces (negative for repulsion, positive for attraction) should improve the description of complex spatial configurations, we propose to extend the force histogram to a panel of forces so as to make it a more complete descriptor. This gives a 2D descriptor that we called "(discrete) Force Banner " and which can be used as input of a classical Convolutional Neural Network (CNN), benefiting from their powerful performances, and reduced into more compact spatial features to use them in another system. As an illustration of its ability to describe spatial configurations, we used it to solve a classification problem aiming to discriminate simple spatial relations, but with variable configuration complexities. Experimental results obtained on datasets of synthetic and natural images with various shapes highlight the interest of this approach, in particular for complex spatial configurations.(c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Although Generative Adversarial Networks (GAN) have shown remarkable performance in image gener-ation, there exist some challenges in instability and convergence speed. During the training, the results of some models display the imbalances of quality within a generated image, in which some defective parts appear compared with other regions. Different from general single global optimization methods, we introduce an adaptive global and local bilevel optimization model (GL-GAN). The model achieves the generation of high-resolution images in a complementary and promoting way, where global optimization is to optimize the whole images and local is only to optimize the low-quality areas. Based on DCGAN, GL-GAN is able to effectively avoid the nature of imbalance by local bilevel optimization, which is ac-complished by first locating low-quality areas and then optimizing them. Moreover, through feature map cues from discriminator output, we propose the adaptive local and global optimization method (Ada-OP) for interactive optimization and observe that it boosts the convergence speed. Compared with the current GAN methods, our model has shown impressive performance on CelebA, Oxford Flowers, CelebA-HQ and LSUN datasets. (c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：The semantic segmentation of building facades is critical for various construction applications, such as urban building reconstruction and damage assessments. As there is a lack of 3D point cloud datasets related to fine-grained building facades, in this work we construct the first large-scale point cloud benchmark dataset for building facade semantic segmentation. In terms of the characteristics of building facade dataset, the existing methods of semantic segmentation cannot fully mine the local neighborhood information of point clouds; therefore, we propose an attention module that learns Dual Local Attention features, called DLA in this paper. The proposed DLA module consists of two blocks, a self-attention block and an attentive pooling block, which both embed an enhanced position encoding block. The DLA module can be easily embedded into various network architectures for point cloud segmentation, naturally resulting in a new 3D semantic segmentation network with an encoder-decoder architecture; we called this network the DLA-Net. Extensive experimental results on our constructed building facade dataset demonstrate that the proposed DLA-Net achieves better performance than the state-of-the-art methods for semantic segmentation. (c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Stroke extraction and matching are critical for structural interpretation based applications of handwrit-ten Chinese characters, such as Chinese character education and calligraphy analysis. Stroke extraction from offline handwritten Chinese characters is difficult because of the missing of temporal information, the multi-stroke structures and the distortion of handwritten shapes. In this paper, we propose a compre-hensive scheme for solving the stroke extraction problem for handwritten Chinese characters. The method consists of three main steps: (1) fully convolutional network (FCN) based skeletonization; (2) query pixel guided stroke extraction; (3) model-based stroke matching. Specifically, based on a recently proposed ar-chitecture of FCN, both the stroke skeletons and cross regions are firstly extracted from the character image by the proposed SkeNet and CrossNet, respectively. Stroke extraction is solved by simulating the human perception that once given a certain pixel from non-cross region of a stroke, the whole stroke containing the pixel can be traced. To realize this idea, we formulate stroke extraction as a problem of pairing and connecting skeleton-wise stroke segments which are adjacent to the same cross region, where the pairing consistency between stroke segments is measured using a PathNet [1]. To reduce the ambiguity of stroke extraction, the extracted candidate strokes are matched with a character model con-sisting of standard strokes by tree search to identify the correct strokes. For verifying the effectiveness of the proposed method, we train and test our models on character images with stroke segmentation an-notations generated from the online handwriting datasets CASIA-OLHWDB and ICDAR13-Online, as well as a dataset of Regularly-Written online handwritten characters (RW-OLHWDB). The experimental results demonstrate the effectiveness of the proposed method and provide several benchmarks. Particularly, the precisions of stroke extraction for ICDAR13-Online and RW-OLHWDB are 89.0% and 94.9%, respectively.(c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：In this study, we present a robust and efficient fingerprint image restoration algorithm using the nonlocal Cahn-Hilliard (CH) equation, which was proposed for modeling the microphase separation of diblock copolymers. We take a small local region embedding the damaged domain and solve the nonlocal CH equation to restore the fingerprint image. A Gauss-Seidel type iterative method, which is efficient and simple to implement, is used. The proposed method has the advantage in that the pixel values in the damaged fingerprint domain can be obtained using the image information from the outside of the damaged fingerprint region. Fingerprint restoration based on adjacent pixel information can ensure the accuracy of the fingerprint information with a low computational cost. Computational experiments demonstrated the superior performance of the proposed fingerprint restoration algorithm. (c) 2021 Elsevier Ltd. All rights reserved.
Ataky, Steve Tsham MpindaKoerich, Alessandro Lameiras
15页
查看更多>>摘要：Texture can be defined as the change of image intensity that forms repetitive patterns resulting from the physical properties of an object's roughness or differences in a reflection on the surface. Considering that texture forms a system of patterns in a non-deterministic way, biodiversity concepts can help its char-acterization from an image. This paper proposes a novel approach to quantify such a complex system of diverse patterns through species diversity, richness, and taxonomic distinctiveness. The proposed ap-proach considers each image channel as a species ecosystem and computes species diversity and richness as well as taxonomic measures to describe the texture. Furthermore, the proposed approach takes ad-vantage of ecological patterns' invariance characteristics to build a permutation, rotation, and translation invariant descriptor. Experimental results on three datasets of natural texture images and two datasets of histopathological images have shown that the proposed texture descriptor has advantages over several texture descriptors and deep methods. (c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：Linear Discriminant Analysis (LDA) assumes that all samples from the same class are independently and identically distributed (i.i.d.). LDA may fail in the cases where the assumption does not hold. Particularly when a class contains several clusters (or subclasses), LDA cannot correctly depict the internal structure as the scatter matrices that LDA relies on are defined at the class level. In order to mitigate the problem, this paper proposes a neighborhood linear discriminant analysis (nLDA) in which the scatter matrices are defined on a neighborhood consisting of reverse nearest neighbors. Thus, the new discriminator does not need an i.i.d. assumption. In addition, the neighborhood can be naturally regarded as the smallest subclass, for which it is easier to be obtained than subclass without resorting to any clustering algorithms. The projected directions are sought to make sure that the within-neighborhood scatter as small as possible and the between-neighborhood scatter as large as possible, simultaneously. The experimental results show that nLDA performs significantly better than previous discriminators, such as LDA, LFDA, ccLDA, LM-NNDA, and l 2 , 1-RLDA.(c) 2021 Elsevier Ltd. All rights reserved.
查看更多>>摘要：To ensure the operational safety and reliability, fault recognition of complex systems is becoming an essential process in industrial systems. However, the existing recognition methods mainly focus on common faults with enough data, which ignore that many faults are lack of samples in engineering practice. Transfer learning can be helpful, but irrelevant knowledge transfer can cause performance degradation, especially in complex systems. To address the above problem, a hierarchy guided transfer learning framework (HGTL) is proposed in this paper for fault recognition with few-shot samples. Firstly, we fuse domain knowledge, label semantics and inter-class distance to calculate the affinity between categories, based on which a category hierarchical tree is constructed by hierarchical clustering. Then, guided by the hierarchical structure, the samples in most similar majority classes are selected from the source domain to pre-train the hierarchical feature learning network (HFN) and extract the transferable fault information. For the fault knowledge extracted from the child nodes of one parent node are similar and can be transferred with each other, so the trained HFN can extract better features of few samples classes with the help of the information from similar faults, and used to address few-shot fault recognition problems. Finally, a dataset of a nuclear power system with 65 categories and the widely used Tennessee Eastman dataset are analyzed respectively via the proposed method, as well as state-of-the-art recognition methods for comparison. The experimental results demonstrate the effectiveness and superiority of the proposed method in fault recognition with few-shot problem. (c) 2021 Elsevier Ltd. All rights reserved.
NSTL
Elsevier