Search

Visual surveillance for autonomous power facility inspection is considered to bethe prominent field of study in the power industry. This research field completely focuses on either object detection or image fusion which lacks the overall consideration. By considering this, a single end-to-end object detection method by incorporating the image fusion named Fast Detection Fusion Network (FDFNet) is proposed in this paper to output qualitative fused images with detection results. The parameters in the FDFnet are greatly reduced by sharing the feature extraction network between image fusion and object detection tasks, due to which a huge reduction in computational complexity is achieved. On this basis, the object detection algorithm performance on various types of power facility images is compared and analyzed. For experimentation purposes, an IR (infrared) and VIS (visible) image acquisition system has also been designed. In addition, a dataset named CVPower with different sets of images for power facility fusion detection is constructed for this research field. Experimental results demonstrate that the proposed method can achieve the mAP of not less than 70%, process 2 frames per second, and produce high qualitative fused images. [ABSTRACT FROM AUTHOR]

Detecting wafer map anomalies is crucial for preventing yield loss in semiconductor fabrication, although intricate patterns and resource-intensive labeled data prerequisites hinder precise deep-learning segmentation. This paper presents an innovative, unsupervised method for detecting pixel-level anomalies in wafer maps. It utilizes an efficient dual attention module with a knowledge distillation network to learn defect distributions without anomalies. Knowledge transfer is achieved by distilling information from a pre-trained teacher into a student network with similar architecture, except an efficient dual attention module is incorporated atop the teacher network’s feature pyramid hierarchies, which enhances feature representation and segmentation across pyramid hierarchies that selectively emphasize relevant and discard irrelevant features by capturing contextual associations in positional and channel dimensions. Furthermore, it enables student networks to acquire an improved knowledge of hierarchical features to identify anomalies across different scales accurately. The dissimilarity in feature pyramids acts as a discriminatory function, predicting the likelihood of an abnormality, resulting in highly accurate pixel-level anomaly detection. Consequently, our proposed method excelled on the WM-811K and MixedWM38 datasets, achieving AUROC, AUPR, AUPRO, and F1-Scores of (99.65%, 99.35%), (97.31%, 92.13%), (90.76%, 84.66%) respectively, alongside an inference speed of 3.204 FPS, showcasing its high precision and efficiency. [ABSTRACT FROM AUTHOR]

Unmanned Aerial Vehicle (UAV) inspection has gradually replaced manual inspection of transmission tower, which produces many images. While it is laborious and time-consuming to manually analyze these images, there are also challenges in automatically detecting small-size defective components such as bolts in transmission tower images, due to problems including complex background, small size, and many similar objects of bolts. In this paper, by virtue of multi-scale features and context information, we propose a deep neural network named Camp-Net (Context Information and Multi-Scale Pyramid Network) to identify bolts defect in transmission tower images. First, multi-scale feature fusion combines deep features and shallow features in convolutional networks to detect small-size bolts. Second, context information fusion puts the information around bolts into the detection network to remove the disturbance of complex background and similar objects. An image dataset containing defective bolts and normal bolts is constructed for model training and testing. Experimental results show that bolts with loose pins and bolts without pins among fittings in transmission tower can be accurately identified with the proposed model. The Average Precision (AP) of defective bolts detection of this model can be 11.4% higher than that of the commonly used high performance model, Faster R-CNN. [ABSTRACT FROM AUTHOR]

In this paper, we present a comprehensive review of the imbalance problems in object detection. To analyze the problems in a systematic manner, we introduce a problem-based taxonomy. Following this taxonomy, we discuss each problem in depth and present a unifying yet critical perspective on the solutions in the literature. In addition, we identify major open issues regarding the existing imbalance problems as well as imbalance problems that have not been discussed before. Moreover, in order to keep our review up to date, we provide an accompanying webpage which catalogs papers addressing imbalance problems, according to our problem-based taxonomy. Researchers can track newer studies on this webpage available at: https://github.com/kemaloksuz/ObjectDetectionImbalance. [ABSTRACT FROM AUTHOR]

Since gait can be utilized to identify individuals from a far distance without their interaction and coordination, recently many gait recognition methods have been proposed. However, due to a real-world scenario of clothing changes, a degradation occurs for most of these methods. Thus in this paper, a more efficient gait recognition method is proposed to address the problem of clothing variances. First, part-based gait features are formulated from two different perspectives, i.e., the separated body parts that are more robust to clothing changes and the estimated human skeleton key-point regions. It is reasonable to formulate such features for cloth-changing gait recognition, because these two perspectives are both less vulnerable to clothing changes. Given that each feature has its own advantages and disadvantages, a more efficient gait feature is generated in this paper by assembling these two features together. Moreover, since local features are more discriminative than global features, in this paper more attention is focused on the local short-range features. Also, unlike most methods, in our method we treat the estimated key-point features as a set of word embeddings, and a transformer encoder is specifically used to learn the dependence of each correlative key-points. The robustness and effectiveness of our proposed method are certified by experiments on CASIA Gait Dataset B, and it has achieved the state-of-the-art performance on this dataset. [ABSTRACT FROM AUTHOR]

Entity resolution (ER) identifies data instances referring to the same real-world entity and has received enormous research attention. In this paper, we examine the task of ER from a broader perspective, with its input extended from textual records, which are conventionally studied in the literature, to other modalities such as check-in sequences, GPS trajectories and surveillance video frames to generate new applications. Our goal in this paper is to design an effective model to uniformly support all these ER applications with different input formats. Technically, we fully exploit the semantic contexts of embedding vectors for the pair of input instances. In particular, we propose an integrated multi-context attention framework that takes into account self-attention, pair-attention and global-attention from three types of context. The idea can be further extended to incorporate attribute attention in order to support structured datasets. We conduct extensive experiments on a diverse class of entity resolutions tasks, including tasks on unstructured, structured and dirty textual records, check-in sequences, GPS trajectories and surveillance video frames. The experimental results verified the effectiveness and generality of our model. When compared with strong baselines in these applications, our model can achieve superior or comparative performance. [ABSTRACT FROM AUTHOR]

This paper presents a novel deep learning-based method that integrates a Convolutional Auto-Encoder (CAE) with support vector machine (SVM) for finger-vein verification. The CAE is used to learn the features from finger-vein images, and the SVM is used to classify finger vein from these learned feature codes. The CAE consists of a finger-vein encoder, which extracts high-level feature representation from raw pixels of the images, and a decoder which outputs reconstruct finger-vein images from high-level feature code. As an effective classifier, SVM is introduced in this paper to classify the feature code which is obtained from CAE. Experiments prove that the proposed deep learning-based approach has superior performance in learning features than traditional method without any prior knowledge, presenting a good potential in the verification of finger vein. [ABSTRACT FROM AUTHOR]

Radio signal recognition has a wide application in future communication systems and the vehicular communication, whose core is the extraction of signal features such as electromagnetic fingerprints. With the rapid development of artificial intelligence technology, deep learning has made amazing breakthroughs in image recognition, speech recognition and other fields. Deep learning is applied to electromagnetic fingerprint extraction in this paper. Firstly, thousands of the downlink aircraft communications addressing and reporting system (ACARS) signals used for communication between civil aircraft and airport tower are collected and generated. Then a pre-transformation network suitable for electromagnetic signals is constructed to convert one-dimensional signals into two-dimensional feature maps, and afterwards the feature maps are input into the convolutional neural network (CNN) for classification. By adopting the attention modules, the classification results were improved by a few percentage points over the baseline with a little cost. The method proposed in this paper achieves an accuracy rate of 94.1% and can obtain the aircraft type in a shorter time than traditional method. Moreover, the robustness of the proposed model in response to additive Gaussian white noise (AWGN) and phase deviation is studied and tested. [ABSTRACT FROM AUTHOR]

The vanilla Few-shot Learning (FSL) learns to build a classifier for a new concept from one or very few target examples, with the general assumption that source and target classes are sampled from the same domain. Recently, the task of Cross-Domain Few-Shot Learning (CD-FSL) aims at tackling the FSL where there is a huge domain shift between the source and target datasets. Extensive efforts on CD-FSL have been made via either directly extending the meta-learning paradigm of vanilla FSL methods, or employing massive unlabeled target data to help learn models. In this paper, we notice that in the CD-FSL task, the few labeled target images have never been explicitly leveraged to inform the model in the training stage. However, such a labeled target example set is very important to bridge the huge domain gap. Critically, this paper advocates a more practical training scenario for CD-FSL. And our key insight is to utilize a few labeled target data to guide the learning of the CD-FSL model. Technically, we propose a novel Generalized Meta-learning based Feature-Disentangled Mixup network, namely GMeta-FDMixup. We make three key contributions of utilizing GMeta-FDMixup to address CD-FSL. Firstly, we present two mixup modules – mixup-P and mixup-M that help facilitate utilizing the unbalanced and disjoint source and target datasets. These two novel modules enable diverse image generation for training the model on the source domain. Secondly, to narrow the domain gap explicitly, we contribute a novel feature disentanglement module that learns to decouple the domain-irrelevant and domain-specific features. By stripping the domain-specific features, we alleviate the negative effects caused by the domain inductive bias. Finally, we repurpose a new contrastive learning module, dubbed ConL. ConL prevents the model from only capturing category-related features via introducing contrastive loss. Thus, the generalization ability on novel categories is improved. Extensive experimental results on two benchmarks show the superiority of our setting and the effectiveness of our method. Code and models will be released. [ABSTRACT FROM AUTHOR]

Detailed routing is one of the most challenging aspects of the physical design process. Many of the violations that occur during the detailed routing stage stem from the placement of the cells. In this paper, we propose a deep learning framework to identify short violations that can occur during detailed routing from a placed netlist. One of the advantages of our technique is that by using the proposed deep learning-based predictor, global routing is no longer required as frequently and hence the total runtime for place and route can be significantly reduced. In this paper, we discuss the proposed framework and the methodology for analyzing the extracted features. The experimental results show that the average sensitivity, specificity, and accuracy of Eh?Predictor is above 90%. In addition, we show that Eh?Predictor is up to 14 times faster than NCTUgr for smaller designs and up to 96 times faster for larger designs. [ABSTRACT FROM AUTHOR]

Virtual metrology (VM) predicts the actual measurement for ongoing semiconductor process. Optical emission spectroscopy (OES) data is often used to build up VM models, since it has a lot of information on process quality. However, it also includes significant redundant information, so it is important how to select only meaningful features. Deep learning (DL) techniques have been very successful in analyzing image data and it is tempting to apply those techniques to the OES data. In this paper, we propose DL configurations specific to OES data that outperform those used for image analysis. Specifically, our proposed method accounts for variable size data, chamber to chamber differences, condition drift due to accumulation, observation data drift due to an accumulation of deposition on a window, and the effects of maintenance. We evaluated our method on a real, mass-production dataset and compared our results with those obtained by using state-of-the-art image analysis DL techniques in the famous contest, ImageNet large-scale visual recognition challenge (ILSVRC). [ABSTRACT FROM AUTHOR]

Large-scale labeled data are generally required to train deep neural networks in order to obtain better performance in visual feature learning from images or videos for computer vision applications. To avoid extensive cost of collecting and annotating large-scale datasets, as a subset of unsupervised learning methods, self-supervised learning methods are proposed to learn general image and video features from large-scale unlabeled data without using any human-annotated labels. This paper provides an extensive review of deep learning-based self-supervised general visual feature learning methods from images or videos. First, the motivation, general pipeline, and terminologies of this field are described. Then the common deep neural network architectures that used for self-supervised learning are summarized. Next, the schema and evaluation metrics of self-supervised learning methods are reviewed followed by the commonly used datasets for images, videos, audios, and 3D data, as well as the existing self-supervised visual feature learning methods. Finally, quantitative performance comparisons of the reviewed methods on benchmark datasets are summarized and discussed for both image and video feature learning. At last, this paper is concluded and lists a set of promising future directions for self-supervised visual feature learning. [ABSTRACT FROM AUTHOR]

Modern methods for counting people in crowded scenes rely on deep networks to estimate people densities in individual images. As such, only very few take advantage of temporal consistency in video sequences, and those that do only impose weak smoothness constraints across consecutive frames. In this paper, we advocate estimating people flows across image locations between consecutive images and inferring the people densities from these flows instead of directly regressing them. This enables us to impose much stronger constraints encoding the conservation of the number of people. As a result, it significantly boosts performance without requiring a more complex architecture. Furthermore, it allows us to exploit the correlation between people flow and optical flow to further improve the results. We also show that leveraging people conservation constraints in both a spatial and temporal manner makes it possible to train a deep crowd counting model in an active learning setting with much fewer annotations. This significantly reduces the annotation cost while still leading to similar performance to the full supervision case. [ABSTRACT FROM AUTHOR]

Due to the diversity of power consumption patterns, the false positive rate (FPR) of data-driven electricity theft detection (ETD) methods is too high to meet practical needs, which severely restricts the engineering application of data-based methods. To reduce FPR of ETD methods based on advanced metering infrastructure (AMI), a deep neural network with low FPR (LFPR-DNN) is proposed in this paper. First, a deep model is constructed based on one-dimensional convolution and residual network, which can automatically extract features from consumption data. Then, a two-stage training scheme is used to train the network. In the first stage, the conventional gradient descent algorithm is used to update the network weights. To minimize the impact of data imbalance on detection performance, focal loss is used. Besides, grid search is used to optimize the hyper-parameters of the model. In the second stage, with FPR as the optimization objective, the particle swarm optimization (PSO) algorithm is used to train the network. Finally, the proposed LFPR-DNN is verified by using the open Irish data set. Compared to other state-of-the-art classifiers, LFPR-DNN has the lowest FPR with 0.29% and the highest AUC with 99.42%. The FPR is reduced by an order of magnitude, which verifies the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Histopathological tissue classification is a simpler way to achieve semantic segmentation for the whole slide images, which can alleviate the requirement of pixel-level dense annotations. Existing works mostly leverage the popular CNN classification backbones in computer vision to achieve histopathological tissue classification. In this paper, we propose a super lightweight plug-and-play module, named Pyramidal Deep-Broad Learning (PDBL), for any well-trained classification backbone to improve the classification performance without a re-training burden. For each patch, we construct a multi-resolution image pyramid to obtain the pyramidal contextual information. For each level in the pyramid, we extract the multi-scale deep-broad features by our proposed Deep-Broad block (DB-block). We equip PDBL in three popular classification backbones, ShuffLeNetV2, EfficientNetb0, and ResNet50 to evaluate the effectiveness and efficiency of our proposed module on two datasets (Kather Multiclass Dataset and the LC25000 Dataset). Experimental results demonstrate the proposed PDBL can steadily improve the tissue-level classification performance for any CNN backbones, especially for the lightweight models when given a small among of training samples (less than 10%). It greatly saves the computational resources and annotation efforts. The source code is available at: https://github.com/linjiatai/PDBL. [ABSTRACT FROM AUTHOR]

Most deep learning models for temporal regression directly output the estimation based on single input images, ignoring the relationships between different images. In this paper, we propose deep relation learning for regression, aiming to learn different relations between a pair of input images. Four non-linear relations are considered: “cumulative relation,” “relative relation,” “maximal relation” and “minimal relation.” These four relations are learned simultaneously from one deep neural network which has two parts: feature extraction and relation regression. We use an efficient convolutional neural network to extract deep features from the pair of input images and apply a Transformer for relation learning. The proposed method is evaluated on a merged dataset with 6,049 subjects with ages of 0–97 years using 5-fold cross-validation for the task of brain age estimation. The experimental results have shown that the proposed method achieved a mean absolute error (MAE) of 2.38 years, which is lower than the MAEs of 8 other state-of-the-art algorithms with statistical significance (p<0.05) in paired T-test (two-side). [ABSTRACT FROM AUTHOR]

Deep learning (DL) techniques have shown promising performance for designing data-driven power system transient stability assessment (TSA) models. However, due to the deep structure of the DL, the resulting model is always a black-box and hard to explain, which hinders its practical adoption by the industry. This paper proposes an interpretable DL-based TSA model to balance the TSA accuracy and transparency. The proposed method combines the strong nonlinear modelling capability of a deep neural network and the interpretability of a decision tree (DT). Through regularizing DL-based model with the average DT path length in the training process, the proposed interpretable DL-based TSA method can visually explain the TSA decision-making process. Simulation results have shown that the proposed method can deliver highly accurate TSA results and interpretable TSA decision-making rules, which can be used for designing preventive control actions. [ABSTRACT FROM AUTHOR]

Deep autoencoder-based approaches have achieved significant improvements on restoring damaged images, yet they still suffer from artifacts due to the inadequate representation and inaccurate regularization of existing features. In this paper, we propose a dual-pyramidal inpainting framework called DPNet to address these two limitations, which seamlessly integrates sufficient feature learning and dynamic regularization within an autoencoder network. Specifically, to exhaustively extract multi-scale features, we adopt layer-wise pyramidal convolution in encoder, which provides an arbitrary combination pool of various receptive fields. Subsequently, to tackle the patch deterioration problem in previous cross-scale non-local schemes, we further propose a Pyramidal Attention Mechanism (PAM) in decoder to acquire finer patches directly from learned layers. Mutually benefited with pyramidal features extraction in encoder, the dissemination space for non-local pixels in our PAM is notably enlarged to pyramidal level, thus significantly benefiting the feature representation. Moreover, to avoid the mask error accumulation in existing works, a dynamic normalization mechanism utilizing the spatial mask information updated in encoder is introduced, which further ensures the feature integrity and consistency. Such a dual-pyramidal structure along with dynamic normalization significantly improve the inpainting quality, outperforming existing competitors. Comprehensive experiments conducted on three benchmark datasets demonstrate that our DPNet performs favorably against the state-of-the-arts. [ABSTRACT FROM AUTHOR]

Text clustering is a critical step in text data analysis and has been extensively studied by the text mining community. Most existing text clustering algorithms are based on the bag-of-words model, which faces the high-dimensional and sparsity problems and ignores text structural and sequence information. Deep learning-based models such as convolutional neural networks and recurrent neural networks regard texts as sequences but lack supervised signals and explainable results. In this paper, we propose a deep feature-based text clustering (DFTC) framework that incorporates pretrained text encoders into text clustering tasks. This model, which is based on sequence representations, breaks the dependency on supervision. The experimental results show that our model outperforms classic text clustering algorithms and the state-of-the-art pretrained language model, i.e., BERT, on almost all the considered datasets. In addition, the explanation of the clustering results is significant for understanding the principles of the deep learning approach. Our proposed clustering framework includes an explanation module that can help users understand the meaning and quality of the clustering results. [ABSTRACT FROM AUTHOR]

Online fault detection of insulators is a necessary requirement for the development of a smart grid, which directly affects the safety and reliability of power system operations. Breakage is an important factor causing insulator abnormalities, and a large number of image samples are often required for training to complete defect identification with the help of AI deep learning. In this paper, we propose a convolutional neural network ensemble implemented using small samples, data augmentation and transfer learning for insulator breakage detection. We used data augmentation to expand the small sample of 200 insulator images by 15 times to 3000 images, built a VGG16 transfer network, pre-trained the VGG16 network convolutional basis weights, and improved the training performance and efficiency of the model with freeze training and fine-tuning training. The training on 2700 images and testing results on 300 images show that the proposed method can achieve an accuracy of 0.9871 for insulator breakage recognition. [ABSTRACT FROM AUTHOR]

This paper presents a deep learning-based method for defect detection and classification of power distribution lines using video analysis. The first stage is dataset preparation in which different videos are recorded by a CoroCam 6D2 camera from distribution lines and labeled based on the defect type and the severity level of the defect. Then, a process is followed in such a way that a limited number of frames are used for processing, and power devices are detected in each frame using Faster R-CNN. The detector performance is improved by increasing the training dataset and changing the training hyper-parameters. Next, an equipment tracking technique is applied through the video frames. In the following, a method based on the classification of power equipment is presented. In this method, by corona color thresholding in all frames and applying a median filter over time, the connected components representing the corona are identified. Then, the frame containing the closest component to the component in the median image is selected. In the selected frame, the area around the component is cut and given to AlexNet or GoogleNet to determine the equipment type and the severity level of defect is determined. Also, the defective equipment phase is determined based on the insulators detected in the video. The proposed method not only performs better than the state-of-the-art but also is a practical method, and with the least dependence on environmental conditions, can automatically identify defects in distribution lines, even in videos containing several possible defective devices. [ABSTRACT FROM AUTHOR]

Current clinical practice relies on clinical history to determine the time since stroke (TSS) onset. Imaging-based determination of acute stroke onset time could provide critical information to clinicians in deciding stroke treatment options, such as thrombolysis. The patients with unknown or unwitnessed TSS are usually excluded from thrombolysis, even if their symptoms began within the therapeutic window. In this paper, we demonstrate a machine learning approach for TSS classification using routinely acquired imaging sequences. We develop imaging features from the magnetic resonance (MR) images and train machine learning models to classify the TSS. We also propose a deep-learning model to extract hidden representations for the MR perfusion-weighted images and demonstrate classification improvement by incorporating these additional deep features. The cross-validation results show that our best classifier achieved an area under the curve of 0.765, with a sensitivity of 0.788 and a negative predictive value of 0.609, outperforming existing methods. We show that the features generated by our deep-learning algorithm correlate with the MR imaging features, and validate the robustness of the model on imaging parameter variations (e.g., year of imaging). This paper advances magnetic resonance imaging analysis one-step-closer to an operational decision support tool for stroke treatment guidance. [ABSTRACT FROM AUTHOR]

A general deep learning (DL) mechanism for a multiple hidden layer feed-forward neural network contains two parts, i.e., 1) an unsupervised greedy layer-wise training and 2) a supervised fine-tuning which is usually an iterative process. Although this mechanism has been demonstrated in many fields to be able to significantly improve the generalization of neural network, there is no clear evidence to show which one of the two parts plays the essential role for the generalization improvement, resulting in an argument within the DL community. Focusing on this argument, this paper proposes a new DL approach to train multilayer feed-forward neural networks. This approach uses restricted Boltzmann machine (RBM) as the layer-wise training and uses the generalized inverse of a matrix as the supervised fine-tuning. Different from the general deep training mechanism like back-propagation (BP), the proposed approach does not need to iteratively tune the weights, and therefore, has many advantages such as quick training, better generalization, and high understandability, etc. Experimentally, the proposed approach demonstrates an excellent performance in comparison with BP-based DL and the traditional training method for multilayer random weight neural networks. To a great extent, this paper demonstrates that the supervised part plays a more important role than the unsupervised part in DL, which provides some new viewpoints for exploring the essence of DL. [ABSTRACT FROM AUTHOR]

The maritime industry widely expects to have autonomous and semiautonomous ships (autoships) in the near future. In order to operate and maintain complex and integrated systems in a safe, efficient, and cost-beneficial manner, autoships will require intelligent Prognostics and Health Management (PHM) systems. Deep learning (DL) is a potential area for this development, as it is rapidly finding applications in a variety of domains, including self-driving cars, smartphones, vision systems, and more recently in PHM applications. This paper introduces and reviews four well-established DL techniques recently applied to various practical PHM problems. The purpose is to support creativity and provide inspiration toward the PHM based on DL in autoships and the maritime industry. This paper discusses benefits, challenges, suggestions, existing problems, and future research opportunities with respect to this significant new technology. [ABSTRACT FROM AUTHOR]

This paper presents an effective diagnosis algorithm for permanent magnet synchronous motors running with an array of faults of varying severity over a wide speed range. The fault diagnosis is based on a current signature analysis. The complete fault motor diagnosis system requires the extraction of features based on the current method and a subsequent method for adding classifications. In this paper, we propose two feature extraction methods: the first involves a classification method that utilizes a wavelet packet transform and the second is a deep 1-D convolution neural network that includes a softmax layer. The experimental results obtained using real-time motor stator current data demonstrate the effectiveness of the proposed methods for real-time monitoring of motor conditions. The results also demonstrate that the proposed methods can effectively diagnose five different motor states, including two different demagnetization fault states and two bearing fault states. [ABSTRACT FROM AUTHOR]

Modern top-performing object detectors depend heavily on backbone networks, whose advances bring consistent performance gains through exploring more effective network structures. In this paper, we propose a novel and flexible backbone framework, namely CBNet, to construct high-performance detectors using existing open-source pre-trained backbones under the pre-training fine-tuning paradigm. In particular, CBNet architecture groups multiple identical backbones, which are connected through composite connections. Specifically, it integrates the high- and low-level features of multiple identical backbone networks and gradually expands the receptive field to more effectively perform object detection. We also propose a better training strategy with auxiliary supervision for CBNet-based detectors. CBNet has strong generalization capabilities for different backbones and head designs of the detector architecture. Without additional pre-training of the composite backbone, CBNet can be adapted to various backbones (i.e., CNN-based vs. Transformer-based) and head designs of most mainstream detectors (i.e., one-stage vs. two-stage, anchor-based vs. anchor-free-based). Experiments provide strong evidence that, compared with simply increasing the depth and width of the network, CBNet introduces a more efficient, effective, and resource-friendly way to build high-performance backbone networks. Particularly, our CB-Swin-L achieves 59.4% box AP and 51.6% mask AP on COCO test-dev under the single-model and single-scale testing protocol, which are significantly better than the state-of-the-art results (i.e., 57.7% box AP and 50.2% mask AP) achieved by Swin-L, while reducing the training time by $6\times $. With multi-scale testing, we push the current best single model result to a new record of 60.1% box AP and 52.3% mask AP without using extra training data. Code is available at https://github.com/VDIGPKU/CBNetV2 . [ABSTRACT FROM AUTHOR]

Recent research advances in salient object detection (SOD) could largely be attributed to ever-stronger multi-scale feature representation empowered by the deep learning technologies. The existing SOD deep models extract multi-scale features via the off-the-shelf encoders and combine them smartly via various delicate decoders. However, the kernel sizes in this commonly-used thread are usually “fixed”. In our new experiments, we have observed that kernels of small size are preferable in scenarios containing tiny salient objects. In contrast, large kernel sizes could perform better for images with large salient objects. Inspired by this observation, we advocate the “dynamic” scale routing (as a brand-new idea) in this paper. It will result in a generic plug-in that could directly fit the existing feature backbone. This paper’s key technical innovations are two-fold. First, instead of using the vanilla convolution with fixed kernel sizes for the encoder design, we propose the dynamic pyramid convolution (DPConv), which dynamically selects the best-suited kernel sizes w.r.t. the given input. Second, we provide a self-adaptive bidirectional decoder design to accommodate the DPConv-based encoder best. The most significant highlight is its capability of routing between feature scales and their dynamic collection, making the inference process scale-aware. As a result, this paper continues to enhance the current SOTA performance. Both the code and dataset are publicly available at https://github.com/wuzhenyubuaa/DPNet. [ABSTRACT FROM AUTHOR]

Most of the existing blind omnidirectional image quality assessment (BOIQA) methods are based on data-driven approach where the end-to-end neural network or deep learning tools are mainly used for feature extraction. However, it usually lacks interpretability and is difficult to discover the perceptual mechanism behind. In this paper, from the perspective of perception modeling, we propose a novel multi-angle projection based BOIQA (MP-BOIQA) method. Considering the omnibearing and near eye display characteristics with head mounted display, multiple color cubemap projection images with respect to different viewpoints are grouped as the color omnidirectional distortion (COD) units so as to simulate the user’s viewing behavior in subjective quality assessment. In the designed multi-angle projection based feature extractor, tensor decomposition is implemented on each COD unit for dimensionality reduction, and piecewise exponential fitting is used to get the distribution of mean subtracted contrast normalized coefficients of the unit’s feature matrices in tensor domain. Finally, the extracted features are pooled with random forest. The experimental results on three omnidirectional image quality datasets show that the MP-BOIQA method can deliver highly competitive performance compared with some representative full-reference quality assessment methods, as well as some state-of-the-art BOIQA methods. [ABSTRACT FROM AUTHOR]

Spatially fine-grained urban flow data is critical for smart city efforts. Though fine-grained information is desirable for applications, it demands much more resources for the underlying storage system compared to coarse-grained data. To bridge the gap between storage efficiency and data utility, in this paper, we aim to infer fine-grained flows throughout a city from their coarse-grained counterparts. This task exhibits two challenges: the spatial correlations between coarse- and fine-grained urban flows, and the complexities of external impacts. To tackle these issues, we develop a model entitled UrbanFM which consists of two major parts: 1) an inference network to generate fine-grained flow distributions from coarse-grained inputs that uses a feature extraction module and a novel distributional upsampling module; 2) a general fusion subnet to further boost the performance by considering the influence of different external factors. This structure provides outstanding effectiveness and efficiency for small scale upsampling. However, the single-pass upsampling used by UrbanFM is insufficient at higher upscaling rates. Therefore, we further present UrbanPy, a cascading model for progressive inference of fine-grained urban flows by decomposing the original tasks into multiple subtasks. Compared to UrbanFM, such an enhanced structure demonstrates favorable performance for larger-scale inference tasks. [ABSTRACT FROM AUTHOR]

Malware developers continually attempt to modify the execution pattern of malicious code hiding it inside apparent normal applications, which makes its detection and classification challenging. This article proposes an ensemble detector, which exploits the capabilities of the main analysis algorithms proposed in the literature designed to offer greater resilience to specific evasion techniques. In particular, the article presents different methods to optimally combine both generic and specialized detectors during the analysis process, which can be used to increase the unpredictability of the detection strategy, as well as improve the detection rate in presence of unknown malware families and provide better detection performance in the absence of a constant re-training of detector needed to cope with the evolution of malware. The paper also presents an alpha-count mechanism that explores how the length of the observation time window can affect the detection accuracy and speed of different combinations of detectors during the malware analysis. An extended experimental campaign has been conducted on both an open-source sandbox and an Android smartphone with different malware datasets. A trade-off among performance, training time, and mean-time-to-detect is presented. Finally, a comparison with other ensemble detectors is also presented. [ABSTRACT FROM AUTHOR]

With the evolution of electronic devices, such as 3D cameras, addressing the challenges of text localization in 3D video (e.g., for indexing) is increasingly drawing the attention of the multimedia and video processing community. Existing methods focus on 2D video and their performance in the presence of the challenges in 3D video, such as shadow areas associated with text and irregularly sized and shaped text, degrades. This paper proposes the first approach that successfully addresses the challenges of 3D video in addition to those of 2D. It employs a number of innovations, among which, the first is the Generalized Gradient Vector Flow (GGVF) for dominant points detection. The second is the Wavefront concept for text candidate point detection from those dominant points. In addition, an Adaptive B-Spline Polygon Curve Network (ABS-Net) is proposed for accurate text localization in 3D videos by constructing tight fitting bounding polygons using text candidate points. Extensive experiments on custom (3D video) and standard datasets (2D video and scene text) show that the proposed method is practical and useful, and overall outperforms existing state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The content based image retrieval aims to find the similar images from a large scale dataset against a query image. Generally, the similarity between the representative features of the query image and dataset images is used to rank the images for retrieval. In early days, various hand designed feature descriptors have been investigated based on the visual cues such as color, texture, shape, etc. that represent the images. However, the deep learning has emerged as a dominating alternative of hand-designed feature engineering from a decade. It learns the features automatically from the data. This paper presents a comprehensive survey of deep learning based developments in the past decade for content based image retrieval. The categorization of existing state-of-the-art methods from different perspectives is also performed for greater understanding of the progress. The taxonomy used in this survey covers different supervision, different networks, different descriptor type and different retrieval type. A performance analysis is also performed using the state-of-the-art methods. The insights are also presented for the benefit of the researchers to observe the progress and to make the best choices. The survey presented in this paper will help in further research progress in image retrieval using deep learning. [ABSTRACT FROM AUTHOR]

Recently, food recognition has received more and more attention in image processing and computer vision for its great potential applications in human health. Most of the existing methods directly extracted deep visual features via convolutional neural networks (CNNs) for food recognition. Such methods ignore the characteristics of food images and are, thus, hard to achieve optimal recognition performance. In contrast to general object recognition, food images typically do not exhibit distinctive spatial arrangement and common semantic patterns. In this paper, we propose a multi-scale multi-view feature aggregation (MSMVFA) scheme for food recognition. MSMVFA can aggregate high-level semantic features, mid-level attribute features, and deep visual features into a unified representation. These three types of features describe the food image from different granularity. Therefore, the aggregated features can capture the semantics of food images with the greatest probability. For that solution, we utilize additional ingredient knowledge to obtain mid-level attribute representation via ingredient-supervised CNNs. High-level semantic features and deep visual features are extracted from class-supervised CNNs. Considering food images do not exhibit distinctive spatial layout in many cases, MSMVFA fuses multi-scale CNN activations for each type of features to make aggregated features more discriminative and invariable to geometrical deformation. Finally, the aggregated features are more robust, comprehensive, and discriminative via two-level fusion, namely multi-scale fusion for each type of features and multi-view aggregation for different types of features. In addition, MSMVFA is general and different deep networks can be easily applied into this scheme. Extensive experiments and evaluations demonstrate that our method achieves state-of-the-art recognition performance on three popular large-scale food benchmark datasets in Top-1 recognition accuracy. Furthermore, we expect this paper will further the agenda of food recognition in the community of image processing and computer vision. [ABSTRACT FROM AUTHOR]

Video-based human action recognition is currently one of the most active research areas in computer vision. Various research studies indicate that the performance of action recognition is highly dependent on the type of features being extracted and how the actions are represented. Since the release of the Kinect camera, a large number of Kinect-based human action recognition techniques have been proposed in the literature. However, there still does not exist a thorough comparison of these Kinect-based techniques under the grouping of feature types, such as handcrafted versus deep learning features and depth-based versus skeleton-based features. In this paper, we analyze and compare 10 recent Kinect-based algorithms for both cross-subject action recognition and cross-view action recognition using six benchmark datasets. In addition, we have implemented and improved some of these techniques and included their variants in the comparison. Our experiments show that the majority of methods perform better on cross-subject action recognition than cross-view action recognition, that the skeleton-based features are more robust for cross-view recognition than the depth-based features, and that the deep learning features are suitable for large datasets. [ABSTRACT FROM AUTHOR]

Remaining useful life (RUL) prediction has been increasingly considered in many industrial fields for the reliability and safety of their systems. As a data analysis tool of deep learning, deep convolutional neural network (CNN) shows great potential for RUL prediction. This paper proposes an intelligent RUL prediction method based on a double-CNN model architecture. Given the powerful feature extraction capability of CNN, the proposed method is fed with original vibration signals with no need to resort to any feature extractor, which can also retain the useful information in maximum. The prediction includes two stages: first, incipient fault point is identified by the first CNN model and a proposed “3/5” principle; then, the second CNN model is constructed for RUL prediction. In practice, RULs of identical components are different from each other, which poses a major challenge in RUL prediction. To overcome this problem, an intermediate reliability variable is first calculated in this paper, instead of directly predicting the RUL value. Then, a mapping algorithm is proposed to map reliability to RUL. To demonstrate the effectiveness of the proposed method, data of four tests of bearing degradation are utilized for RUL prediction. Compared with state-of-the-art methods, the proposed method shows higher prediction accuracy and robustness. The prediction results and evaluation indexes demonstrated the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Automatic target recognition (ATR) is an important part for many computer vision applications. Despite the extensive research which has been carried out in this area for many years, there is no ATR system which performs well on all applications. Recently, different object recognition frameworks have been proposed which yield a high performance in baseline databases. However, our experiments showed that they can fail in real-world scenarios, when dealing with a limited number of data samples. In this paper, we propose a new ATR system, based on deep convolutional neural network (DCNN), to detect the targets in forward looking infrared (FLIR) scenes and recognize their classes. In our proposed ATR framework, a fully convolutional network is trained to map the input FLIR imagery data to a fixed stride correspondingly-sized target score map. The potential targets are identified by applying a threshold on the target score map. Finally, the corresponding regions centered at these target points are fed to a DCNN to classify them into different target types while at the same time rejecting the false alarms. The proposed architecture achieves a significantly better performance in comparison with that of the state-of-the-art methods on two large FLIR image databases. [ABSTRACT FROM AUTHOR]

With the development of video surveillance in public safety field, there is an increasing research on person re-identification (re-id). In this paper, we address the image-to-video person re-id, in which the probe is an image and the gallery is consists of videos captured by nonoverlapping cameras. Compared with image, video sequence contains more temporal information that can be explored to improve the performance of re-identification system. However, it is challenging to model temporal information in the matching process of image-to-video person re-id. In this paper, we proposed a novel temporally memorized similarity learning neural network for this problem. In specific, the proposed network mainly consisted of two parts, including feature representation sub-network and similarity sub-network. In the first part, we adopted a convolutional neural network (CNN) to extract features from the input image. Given a video sequence of a person, features were first extracted from each its frame by using CNN and further forward to a long shot term memory (LSTM) network to encode the temporal information of video sequence. The outputs of LSTM were concatenated together as the feature vector of video sequences. Finally, the feature vectors of probe image and the video sequence were further forward to the similarity sub-network for distance metric learning. In the proposed framework, the feature representation and the similarity metric learning can be learned and optimized simultaneously. We evaluated the proposed framework on three public person re-id data sets, and the experimental results showed that the proposed approach is effective for the image-to-video person re-id. [ABSTRACT FROM AUTHOR]

In this paper, we propose a spectral–spatial unified network (SSUN) with an end-to-end architecture for the hyperspectral image (HSI) classification. Different from traditional spectral–spatial classification frameworks where the spectral feature extraction (FE), spatial FE, and classifier training are separated, these processes are integrated into a unified network in our model. In this way, both FE and classifier training will share a uniform objective function and all the parameters in the network can be optimized at the same time. In the implementation of the SSUN, we propose a band grouping-based long short-term memory model and a multiscale convolutional neural network as the spectral and spatial feature extractors, respectively. In the experiments, three benchmark HSIs are utilized to evaluate the performance of the proposed method. The experimental results demonstrate that the SSUN can yield a competitive performance compared with existing methods. [ABSTRACT FROM AUTHOR]

Recent studies have shown that a Deep Convolutional Neural Network (DCNN) trained on a large image dataset can be used as a universal image descriptor and that doing so leads to impressive performance for a variety of image recognition tasks. Most of these studies adopt activations from a single DCNN layer, usually a fully-connected layer, as the image representation. In this paper, we proposed a novel way to extract image representations from two consecutive convolutional layers: one layer is used for local feature extraction and the other serves as guidance to pool the extracted features. By taking different viewpoints of convolutional layers, we further develop two schemes to realize this idea. The first directly uses convolutional layers from a DCNN. The second applies the pre-trained CNN on densely sampled image regions and treats the fully-connected activations of each image region as a convolutional layer's feature activations. We then train another convolutional layer on top of that as the pooling-guidance convolutional layer. By applying our method to three popular visual classification tasks, we find that our first scheme tends to perform better on applications which demand strong discrimination on lower-level visual patterns while the latter excels in cases that require discrimination on category-level patterns. Overall, the proposed method achieves superior performance over existing approaches for extracting image representations from a DCNN. In addition, we apply cross-layer pooling to the problem of image retrieval and propose schemes to reduce the computational cost. Experimental results suggest that the proposed method achieves promising results for the image retrieval task. [ABSTRACT FROM AUTHOR]

Smoke detection in video is a challenging task because of the irregular shape of smoke, its complex motion state, which is affected by temperature, wind and other external factors, and background disturbances. Pixel-based foreground modeling method is a crucial step in many smoke detection systems and can be applied to efficiently focus on a certain object or a specific region to detect movements or anomalies. In video analysis, it is a natural idea to move the focus from the pixel-level foreground to the feature-level foreground. In this paper, the feature foreground is generated by the middle layer of a convolutional neural network (CNN) to guide the temporal modeling process for smoke objects. A novel temporal module called the Feature Foreground Module (FFM) is proposed to boost learning of a smoke temporal representation. Consider the problem of smoke analysis in video, we present a novel unifying approach, named an enhanced feature foreground network (EFFNet), that performs both smoke source prediction and detection. Efficient branch networks are designed in EFFNet, to predict the source mask and bounding boxes of smoke plumes in video. To the best of our knowledge, this is the first paper to study the source of smoke using deep learning methods. Finally, experiments on a realistic smoke dataset and a public dataset show that EFFNet method performs much better than do previous state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Wafer bin maps contain vital information that helps semiconductor manufacturers to identify the root causes and defect pattern failures in wafers. Conventional manual inspection techniques in inspecting these failures are labour intensive and cause prolonged production cycle time. Therefore, automatic inspection techniques can solve this problem. This paper proposes a deep learning approach based on deep convolutional generative adversarial network (DCGAN) and a new Capsule Network (WaferCaps). DCGAN was used to upsample the original dataset and therefore increase the data used for training and balance the classes at the same time. While WaferCaps was proposed to classify the defect patterns according to eight classes. The performance of our proposed DCGAN and WaferCaps was compared with different deep learning models such as the original Capsule Network (CapsNet), CNN, and MLP. In all of our experiment, WM-811K dataset was used for the data upsampling and training. The proposed approach has shown an effective performance in generating new synthetic data and classify them with training accuracy of 99.59%, validation accuracy of 97.53% and test accuracy of 91.4%. [ABSTRACT FROM AUTHOR]

Online services are playing critical roles in almost all aspects of users’ life. Users usually have multiple online identities (IDs) in different online services. In order to fuse the separated user data in multiple services for better business intelligence, it is critical for service providers to link online IDs belonging to the same user. On the other hand, the popularity of mobile networks and GPS-equipped smart devices have provided a generic way to link IDs, i.e., utilizing the mobility traces of IDs. However, linking IDs based on their mobility traces has been a challenging problem due to the highly heterogeneous, incomplete and noisy mobility data across services. In this paper, we propose DPLink, an end-to-end deep learning based framework, to complete the user identity linkage task for heterogeneous mobility data collected from different services with different properties. DPLink is made up by a feature extractor including a location encoder and a trajectory encoder to extract representative features from trajectory and a comparator to compare and decide whether to link two trajectories as the same user. Particularly, we propose a pre-training strategy with a simple task to train the DPLink model to overcome the training difficulties introduced by the highly heterogeneous nature of different source mobility data. Besides, we introduce a multi-modal embedding network and a co-attention mechanism in DPLink to deal with the low-quality problem of mobility data. By conducting extensive experiments on two real-life ground-truth mobility datasets with eight baselines, we demonstrate that DPLink outperforms the state-of-the-art solutions by more than 15 percent in terms of hit-precision. Moreover, it is expandable to add external geographical context data and works stably with heterogeneous noisy mobility traces. [ABSTRACT FROM AUTHOR]

Surface defect detection plays an important role in manufacturing and has aroused widespread interests. However, it is challenging as defects are highly similar to non-defects. To address this issue, this paper proposes a Region of Interest Attention (RoIA) network based on deep learning for automatically identifying surface defects. It consists of three parts: multi-level feature preservation (MFP) module, region proposal attention (RPA) module, and skip dense connection detection (SDCD) ones, where MFP is designed to differentiate defect features and texture information by feature reserved block, RPA is developed to locate the position of the defects by capturing global and local context information, and SDCD is proposed to better predict defect categories by propagating the fine-grained details from low-level feature map to high-level one. Experimental results conducted on three public datasets (e.g., NEU-DET, DAGM and Magnetic-Tile) demonstrate that the proposed method can significantly improve the detection performance than state-of-the-art ones and achieve an average defect detection accuracy of 99.49%. [ABSTRACT FROM AUTHOR]

Deep convolutional neural networks (CNNs) have emerged as a new paradigm for Mammogram diagnosis. Contemporary CNN-based computer-aided-diagnosis systems (CADs) for breast cancer directly extract latent features from input mammogram image and ignore the importance of morphological features. In this paper, we introduce a novel end-to-end deep learning framework for mammogram image processing, which computes mass segmentation and simultaneously predicts diagnosis results. Specifically, our method is constructed in a dual-path architecture that solves the mapping in a dual-problem manner, with an additional consideration of important shape and boundary knowledge. One path, called the Locality Preserving Learner (LPL), is devoted to hierarchically extracting and exploiting intrinsic features of the input. Whereas the other path, called the Conditional Graph Learner (CGL), focuses on generating geometrical features via modeling pixel-wise image to mask correlations. By integrating the two learners, both the cancer semantics and cancer representations are well learned, and the component learning paths in return complement each other, contributing an improvement to the mass segmentation and cancer classification problem at the same time. In addition, by integrating an automatic detection set-up, the DualCoreNet achieves fully automatic breast cancer diagnosis practically. Experimental results show that in benchmark DDSM dataset, DualCoreNet has outperformed other related works in both segmentation and classification tasks, achieving 92.27% DI coefficient and 0.85 AUC score. In another benchmark INbreast dataset, DualCoreNet achieves the best mammography segmentation (93.69% DI coefficient) and competitive classification performance (0.93 AUC score). [ABSTRACT FROM AUTHOR]

Portfolio Selection is an important real-world financial task and has attracted extensive attention in artificial intelligence communities. This task, however, has two main difficulties: (i) the non-stationary price series and complex asset correlations make the learning of feature representation very hard; (ii) the practicality principle in financial markets requires controlling both transaction and risk costs. Most existing methods adopt handcraft features and/or consider no constraints for the costs, which may make them perform unsatisfactorily and fail to control both costs in practice. In this paper, we propose a cost-sensitive portfolio selection method with deep reinforcement learning. Specifically, a novel two-stream portfolio policy network is devised to extract both price series patterns and asset correlations, while a new cost-sensitive reward function is developed to maximize the accumulated return and constrain both costs via reinforcement learning. We theoretically analyze the near-optimality of the proposed reward, which shows that the growth rate of the policy regarding this reward function can approach the theoretical optimum. We also empirically evaluate the proposed method on real-world datasets. Promising results demonstrate the effectiveness and superiority of the proposed method in terms of profitability, cost-sensitivity and representation abilities. [ABSTRACT FROM AUTHOR]

Fault diagnosis of insulators in aerial images is an essential task of power line inspection to maintain the reliability, safety, and sustainability of power transmission. This paper develops a novel method for intelligent diagnosis of electrical insulators based on deep learning, termed Box-Point Detector, which consists of a deep convolutional neural network followed by two parallel branches of convolutional heads. These two branches are utilized to locate the fault region and estimate insulator endpoints, which presents a new representation for insulator faults. Endpoints of the faulty insulator string can provide detailed and correlative information for enhancing the diagnosis capability of component-dependent faults that occur on component bodies. The proposed Box-Point Detector implements all predictions including region and endpoint into one network thus forms an efficient end-to-end structure, and adopts a smaller downsampling ratio to generate high resolution feature-maps in order to preserve more original information for small size faults. Experimental results indicate that Box-Point Detector can accurately diagnose high-voltage insulator faults in real-time under various conditions. Compared with some previous works using Faster R-CNN, SSD, and cascading network, our Box-Point Detector shows more competitively capabilities with high accuracy and robustness. [ABSTRACT FROM AUTHOR]

In this paper, a novel Deep Multi-view Joint Clustering (DMJC) framework is proposed, where multiple deep embedded features, multi-view fusion mechanism, and clustering assignments can be learned simultaneously. Through the joint learning strategy, the clustering-friendly multi-view features and useful multi-view complementary information can be exploited effectively to improve the clustering performance. Under the proposed joint learning framework, we design two ingenious variants of deep multi-view joint clustering models, whose multi-view fusion is implemented by two kinds of simple yet effective schemes. The first model, called DMJC-S, performs multi-view fusion in an implicit way via a novel multi-view soft assignment distribution. The second model, termed DMJC-T, defines a novel multi-view auxiliary target distribution to conduct the multi-view fusion explicitly. Both DMJC-S and DMJC-T are optimized under a KL divergence objective. Experiments on eight challenging image datasets demonstrate the superiority of both DMJC-S and DMJC-T over single/multi-view baselines and the state-of-the-art multi-view clustering methods, which proves the effectiveness of the proposed DMJC framework. To the best of our knowledge, this is the first work to model the multi-view clustering in a deep joint framework, which will provide a meaningful thinking in unsupervised multi-view learning. [ABSTRACT FROM AUTHOR]

In vehicle trajectory prediction, the difficulty in modeling the interaction relationship between vehicles lies in constructing the interaction structure between the vehicles in the traffic scene. Majority of existing models only focus on the interaction between the historical trajectory of the vehicle and the surrounding vehicles in the spatial domain, and do not pay attention to the interaction between the vehicle and the non-Euclidean correlation structure (graph structure) that exists in the environment. In order to overcome the deficiencies in the existing models, this paper proposes the Environment-Attention Network model (EA-Net) to obtain the full interactive information between the vehicle and its driving environment. In the proposed model, a new type of parallel structure consisting of Graph Attention network (GAT) and Convolutional social pooling containing Squeeze-and-Extraction mechanism (SE-CS), is constructed as the environmental feature extraction module and embedded in LSTM encoder-decoder. This structure solves the limitation of the dimension and structure influence when constructing the interaction model between the vehicle and the surrounding environment, making the extracted feature information comprehensive and effective. The prediction accuracy of the model with the RMSE loss function is tested on two public datasets– NGSIM and highD, and compared with several state-of-the-art trajectory prediction algorithm models. The results show that the prediction accuracy of the proposed Environment-Attention Network in the two datasets is more than 20% higher than that of the single structure model, which indicates that the proposed model proposed has superior performance and better adaptability to different traffic environments compared with the existing models. [ABSTRACT FROM AUTHOR]