Biblio

At present, pavement crack detection mainly depends on manual survey and semi-automatic detection. In the process of damage detection, it will inevitably be subject to the subjective influence of inspectors and require a lot of identification time. Therefore, this paper proposes the research and implementation of artificial intelligence real-time recognition method of crack edge based on zynq, which combines edge calculation technology with deep learning, The improved ipd-yolo target detection network is deployed on the zynq zu2cg edge computing development platform. The mobilenetv3 feature extraction network is used to replace the cspdarknet53 feature extraction network in yolov4, and the deep separable convolution is used to replace the conventional convolution. Combined with the advantages of the deep neural network in the cloud and edge computing, the rock fracture detection oriented to the edge computing scene is realized. The experimental results show that the accuracy of the network on the PID data set The recall rate and F1 score have been improved to better meet the requirements of real-time identification of rock fractures.

The object detection tasks based on edge computing have received great attention. A common concern hasn't been addressed is that edge may be unreliable and uploads the incorrect data to cloud. Existing works focus on the consistency of the transmitted data by edge. However, in cases when the inputs and the outputs are inherently different, the authenticity of data processing has not been addressed. In this paper, we first simply model the tampering detection. Then, bases on the feature insertion and game theory, the tampering detection and economic incentives mechanism (TDEI) is proposed. In tampering detection, terminal negotiates a set of features with cloud and inserts them into the raw data, after the cloud determines whether the results from edge contain the relevant information. The honesty incentives employs game theory to instill the distrust among different edges, preventing them from colluding and thwarting the tampering detection. Meanwhile, the subjectivity of nodes is also considered. TDEI distributes the tampering detection to all edges and realizes the self-detection of edge results. Experimental results based on the KITTI dataset, show that the accuracy of detection is 95% and 80%, when terminal's additional overhead is smaller than 30% for image and 20% for video, respectively. The interference ratios of TDEI to raw data are about 16% for video and 0% for image, respectively. Finally, we discuss the advantage and scalability of TDEI.

Aiming at the single hopping strategy in the terminal information hopping active defense technology, a variety of heterogeneous hopping modes are introduced into the terminal information hopping system, the definition of the terminal information is expanded, and the adaptive adjustment of the hopping strategy is given. A network adversarial training simulation system is researched and designed, and related subsystems are discussed from the perspective of key technologies and their implementation, including interactive adversarial training simulation system, adversarial training simulation support software system, adversarial training simulation evaluation system and adversarial training Mock Repository. The system can provide a good environment for network confrontation theory research and network confrontation training simulation, which is of great significance.

Co-salient object detection (CoSOD) has recently achieved significant progress and played a key role in retrieval-related tasks. However, it inevitably poses an entirely new safety and security issue, i.e., highly personal and sensitive content can potentially be extracting by powerful CoSOD methods. In this paper, we address this problem from the perspective of adversarial attacks and identify a novel task: adversarial co-saliency attack. Specially, given an image selected from a group of images containing some common and salient objects, we aim to generate an adversarial version that can mislead CoSOD methods to predict incorrect co-salient regions. Note that, compared with general white-box adversarial attacks for classification, this new task faces two additional challenges: (1) low success rate due to the diverse appearance of images in the group; (2) low transferability across CoSOD methods due to the considerable difference between CoSOD pipelines. To address these challenges, we propose the very first blackbox joint adversarial exposure and noise attack (Jadena), where we jointly and locally tune the exposure and additive perturbations of the image according to a newly designed high-feature-level contrast-sensitive loss function. Our method, without any information on the state-of-the-art CoSOD methods, leads to significant performance degradation on various co-saliency detection datasets and makes the co-salient objects undetectable. This can have strong practical benefits in properly securing the large number of personal photos currently shared on the Internet. Moreover, our method is potential to be utilized as a metric for evaluating the robustness of CoSOD methods.

Aim: Object Detection is one of the latest topics in today’s world for detection of real time objects using Deep Belief Networks. Methods & Materials: Real-Time Object Detection is performed using Deep Belief Networks (N=24) over Convolutional Neural Networks (N=24) with the split size of training and testing dataset 70% and 30% respectively. Results: Deep Belief Networks has significantly better accuracy (81.2%) compared to Convolutional Neural Networks (47.7%) and attained significance value of p = 0.083. Conclusion: Deep Belief Networks achieved significantly better object detection than Convolutional Neural Networks for identifying real-time objects in traffic surveillance.

The ongoing COVID-19 virus pandemic has resulted in a global tragedy due to its lethal spread. The population's vulnerability grows as a result of a lack of effective helping agents and vaccines against the virus. The spread of viruses can be mitigated by minimizing close connections between people. Social distancing is a critical containment tool for COVID-19 prevention. In this paper, the social distancing violations that are being made by the people when they are in public places are detected. As per CDC (Centers for Disease Control and Prevention) minimum distance that should be maintained by people is 2-3 meters to prevent the spread of COVID- 19, the proposed tool will be used to detect the people who are maintaining less than 2-3 meters of distance between themselves and record them as a violation. As a result, the goal of this work is to develop a deep learning-based system for object detection and tracking models in social distancing detection. For object detection models, You Only Look Once, Version 3 (YOLO v3) is used in conjunction with deep sort algorithms to balance speed and accuracy. To recognize persons in video segments, the approach applies the YOLOv3 object recognition paradigm. An efficient computer vision-based approach centered on legitimate continuous tracking of individuals is presented to determine supportive social distancing in public locations by creating a model to generate a supportive climate that contributes to public safety and detect violations through camera.

Trust is a cognitive ability that can be dependent on behavioral consistency. In this paper, a partially observable Markov Decision Process (POMDP)-based computational robot-human trust model is proposed for hand-over tasks in human-robot collaborative contexts. The robot's trust in its human partner is evaluated based on the human behavior estimates and object detection during the hand-over task. The human-robot hand-over process is parameterized as a partially observable Markov Decision Process. The proposed approach is verified in real-world human-robot collaborative tasks. Results show that our approach can be successfully applied to human-robot hand-over tasks to achieve high efficiency, reduce redundant robot movements, and realize predictability and mutual understanding of the task.

In defense and security applications, detection of moving target direction is as important as the target detection and/or target classification. In this study, a methodology for the detection of different mobile targets as approaching or receding was proposed for ground surveillance radar data, and convolutional neural networks (CNN) based on transfer learning were employed for this purpose. In order to improve the classification performance, the use of two key concepts, namely Deep Convolutional Generative Adversarial Network (DCGAN) and decision fusion, has been proposed. With DCGAN, the number of limited available data used for training was increased, thus creating a bigger training dataset with identical distribution to the original data for both moving directions. This generated synthetic data was then used along with the original training data to train three different pre-trained deep convolutional networks. Finally, the classification results obtained from these networks were combined with decision fusion approach. In order to evaluate the performance of the proposed method, publicly available RadEch dataset consisting of eight ground target classes was utilized. Based on the experimental results, it was observed that the combined use of the proposed DCGAN and decision fusion methods increased the detection accuracy of moving target for person, vehicle, group of person and all target groups, by 13.63%, 10.01%, 14.82% and 8.62%, respectively.

With 3.78 billion social media users worldwide in 2021 (48% of the human population), almost 3 billion images are shared daily. At the same time, a consistent evolution of smartphone cameras has led to a photography explosion with 85% of all new pictures being captured using smartphones. However, lately, there has been an increased discussion of privacy concerns when a person being photographed is unaware of the picture being taken or has reservations about the same being shared. These privacy violations are amplified for people with disabilities, who may find it challenging to raise dissent even if they are aware. Such unauthorized image captures may also be misused to gain sympathy by third-party organizations, leading to a privacy breach. Privacy for people with disabilities has so far received comparatively less attention from the AI community. This motivates us to work towards a solution to generate privacy-conscious cues for raising awareness in smartphone users of any sensitivity in their viewfinder content. To this end, we introduce PrivPAS (A real time Privacy-Preserving AI System) a novel framework to identify sensitive content. Additionally, we curate and annotate a dataset to identify and localize accessibility markers and classify whether an image is sensitive to a featured subject with a disability. We demonstrate that the proposed lightweight architecture, with a memory footprint of a mere 8.49MB, achieves a high mAP of 89.52% on resource-constrained devices. Furthermore, our pipeline, trained on face anonymized data. achieves an F1-score of 73.1%.

In order to solve the problem of large data volume brought by the traditional Nyquist sampling theorem in radar signal detection, a compressive detection (CD) model based on compressed sensing (CS) theory is proposed by analyzing the sparsity of the radar target in the range domain. The lower sampling rate completes the compressive sampling of the radar signal on the range field. On this basis, the two-dimensional distribution of the Doppler unit is established by moving target detention moving target detention (MTD), and the detection of the target is achieved with the two-dimensional constant false alarm rate (2D-CFAR) detection algorithm. The simulation experiment results prove that the algorithm can effectively detect the target without the need for reconstruction signals, and has good detection performance.

Most of the studies of the existing object detection models are studies to better detect the objects to be detected. The problem of false detection of objects that should not be detected is not considered. When an object detection model that does not take this problem into account is applied to an industrial field close to humans, false detection can lead to a dangerous situation that greatly interferes with human life. To solve this false detection problem, this paper proposes a method of fine-tuning the backbone neural network model of the object detection model using the Outlier Exposure method and applying the class-specific uncertainty constant to the confidence score to detect the object.

Enhancement of the vanilla generative adversarial network (GAN) to preserve data variability in the presence of real world noise is of paramount significance in deep learning. In this study, we proposed a new distance metric of cosine distance in the framework of optimal transport (OT), and presented and validated a convolutional neural network (CNN) based GAN framework. In comparison with state-of-the-art methods based on Graphics Processing Units (GPU), the proposed framework could maintain the data diversity and quality best in terms of inception score (IS), Fréchet inception distance (FID) and enhancing the classification network of bone age, and is robust to noise degradation. The proposed framework is independent of hardware and thus could also be extended to more advanced hardware such as specialized Tensor Processing Units (TPU), and could be a potential built-in component of a general deep learning networks for such applications as image classification, segmentation, registration, and object detection.

Currently, more and more malicious insiders are making threats, and the detection of insider threats is becoming more challenging. The malicious insider often uses legitimate access privileges and mimic normal behaviors to evade detection, which is difficult to be detected via using traditional defensive solutions. In this paper, we propose DeepMIT, a malicious insider threat detection framework, which utilizes Recurrent Neural Network (RNN) to model user behaviors as time sequences and predict the probabilities of anomalies. This framework allows DeepMIT to continue learning, and the detections are made in real time, that is, the anomaly alerts are output as rapidly as data input. Also, our framework conducts further insight of the anomaly scores and provides the contributions to the scores and, thus, significantly helps the operators to understand anomaly scores and take further steps quickly(e.g. Block insider's activity). In addition, DeepMIT utilizes user-attributes (e.g. the personality of the user, the role of the user) as categorical features to identify the user's truly typical behavior, which help detect malicious insiders who mimic normal behaviors. Extensive experimental evaluations over a public insider threat dataset CERT (version 6.2) have demonstrated that DeepMIT has outperformed other existing malicious insider threat solutions.

The present study proposes a method for detecting objects with a high recall rate for human-supported video annotation. In recent years, automatic annotation techniques such as object detection and tracking have become more powerful; however, detection and tracking of occluded objects, small objects, and blurred objects are still difficult. In order to annotate such objects, manual annotation is inevitably required. For this reason, we envision a human-supported video annotation framework in which over-detected objects (i.e., false positives) are allowed to minimize oversight (i.e., false negatives) in automatic annotation and then the over-detected objects are removed manually. This study attempts to achieve human-in-the-loop object detection with an emphasis on suppressing the oversight for the former stage of processing in the aforementioned annotation framework: bi-directional deep SORT is proposed to reliably capture missed objects and annotation-free segment identification (AFSID) is proposed to identify video frames in which manual annotation is not required. These methods are reinforced each other, yielding an increase in the detection rate while reducing the burden of human intervention. Experimental comparisons using a pedestrian video dataset demonstrated that bi-directional deep SORT with AFSID was successful in capturing object candidates with a higher recall rate over the existing deep SORT while reducing the cost of manpower compared to manual annotation at regular intervals.

X-ray baggage security screening is a demanding task for aviation and rail transit security; automatic prohibited item detection in X-ray baggage images can help reduce the work of inspectors. However, as many items are placed too close to each other in the baggages, it is difficult to fully trust the detection results of intelligent prohibited item detection algorithms. In this paper, a human-in-the-loop baggage inspection framework is proposed. The proposed framework utilizes the deep-learning-based algorithm for prohibited item detection to find suspicious items in X-ray baggage images, and select manual examination when the detection algorithm cannot determine whether the baggage is dangerous or safe. The advantages of proposed inspection process include: online to capture new sample images for training incrementally prohibited item detection model, and augmented prohibited item detection intelligence with human-computer collaboration. The preliminary experimental results show, human-in-the-loop process by combining cognitive capabilities of human inspector with the intelligent algorithms capabilities, can greatly improve the efficiency of in-baggage security screening.

The use of video surveillance (VS) has grown significantly using the internet as a platform. Thus security issues on such videos must be addressed. Video frames can have multiple objects and various features over video length. Moving object detection (MOD) and real-time tracking requires security strategies designed to protect videos. This paper is proposed to design an asymmetric encryption method (RSA). The paper has contributed in two stages. In the first phase the fast video segmentation method based on a global variable threshold is designed to facilitate MOD. Later in second pass the RSA-based encryption is used to maintain the efficiency of the object detection. The secure key generation method is demonstrated. The performances of two global thresholds are demonstrated and compared under the encrypted video data. It is found that that method is very effective in finding objects under the context of video surveillance in real time.

Autonomous vehicles increasingly exploit computer-vision-based object detection systems to perceive environments and make critical driving decisions. To increase the quality of images, image stabilizers with inertial sensors are added to alleviate image blurring caused by camera jitters. However, such a trend opens a new attack surface. This paper identifies a system-level vulnerability resulting from the combination of the emerging image stabilizer hardware susceptible to acoustic manipulation and the object detection algorithms subject to adversarial examples. By emitting deliberately designed acoustic signals, an adversary can control the output of an inertial sensor, which triggers unnecessary motion compensation and results in a blurred image, even if the camera is stable. The blurred images can then induce object misclassification affecting safety-critical decision making. We model the feasibility of such acoustic manipulation and design an attack framework that can accomplish three types of attacks, i.e., hiding, creating, and altering objects. Evaluation results demonstrate the effectiveness of our attacks against four academic object detectors (YOLO V3/V4/V5 and Fast R-CNN), and one commercial detector (Apollo). We further introduce the concept of AMpLe attacks, a new class of system-level security vulnerabilities resulting from a combination of adversarial machine learning and physics-based injection of information-carrying signals into hardware.

The emergence of the 5G network has boosted the advancements in the field of the internet of things (IoT) and edge/cloud computing. We present a novel architecture to detect fire in indoor and outdoor environments, dubbed as EAC-FD, an abbreviation of edge and cloud-based fire detection. Compared with existing frameworks, ours is lightweight, secure, cost-effective, and reliable. It utilizes a hybrid edge and cloud computing framework with Intel neural compute stick 2 (NCS2) accelerator is for inference in real-time with Raspberry Pi 3B as an edge device. Our fire detection model runs on the edge device while also capable of cloud computing for more robust analysis making it a secure system. We compare different versions of SSD-MobileNet architectures with ours suitable for low-end devices. The fire detection model shows a good balance between computational cost frames per second (FPS) and accuracy.

Nowadays, video surveillance cameras are widely used in many smart city applications for ensuring road safety. We can use video data from them to solve such tasks as traffic management, driving control, environmental monitoring, etc. Most of these applications are based on object recognition and tracking algorithms. However, the video image quality is not always meet the requirements of such algorithms due to the influence of different external factors. A variety of adverse weather conditions produce noise on the images, which often makes it difficult to detect objects correctly. Lately, deep learning methods show good results in image processing, including denoising tasks. This work is devoted to the study of using these methods for image quality enhancement in difficult weather conditions such as snow, rain, fog. Different deep learning techniques were evaluated in terms of their impact on the quality of object detection/recognition. Finally, the system for automatic image denoising was developed.

Due to the continuous improvement of deep learning, saliency object detection based on deep learning has been a hot topic in computational vision. The Fully Convolutional Neural Network (FCNS) has become the mainstream method in salient target measurement. In this article, we propose a new end-to-end multi-level feature fusion module(MCFB), success-fully achieving the goal of extracting rich multi-scale global information by integrating semantic and detailed information. In our module, we obtain different levels of feature maps through convolution, and then cascade the different levels of feature maps, fully considering our global information, and get a rough saliency image. We also propose an optimization module upon our base module to further optimize the feature map. To obtain a clearer boundary, we use a self-defined loss function to optimize the learning process, which includes the Intersection-over-Union (IoU) losses, Binary Cross-Entropy (BCE), and Structural Similarity (SSIM). The module can extract global information to a greater extent while obtaining clearer boundaries. Compared with some existing representative methods, this method has achieved good results.

Social conflicts appearing in the media are increasing public awareness about security issues, resulting in a higher demand of more exhaustive environment monitoring methods. Automatic video surveillance systems are a powerful assistance to public and private security agents. Since the arrival of deep learning, object detection and classification systems have experienced a large improvement in both accuracy and versatility. However, deep learning-based object detection and classification systems often require expensive GPU-based hardware to work properly. This paper presents a novel deep learning-based foreground anomalous object detection system for video streams supplied by panoramic cameras, specially designed to build power efficient video surveillance systems. The system optimises the process of searching for anomalous objects through a new potential detection generator managed by three different multivariant homoscedastic distributions. Experimental results obtained after its deployment in a Jetson TX2 board attest the good performance of the system, postulating it as a solvent approach to power saving video surveillance systems.

Style transfer means using a neural network to extract the content of one image and the style of the other image. The two are combined to get the final result, broadly applied in social communication, animation production, entertainment items. Using style transfer, users can share and exchange images; painters can create specific art styles more readily with less creation cost and production time. Therefore, style transfer is widely concerned recently due to its various and valuable applications. In the past few years, the paper reviews style transfer and chooses three representative works to analyze in detail and contrast with each other, including StyleGAN, CycleGAN, and TL-GAN. Moreover, what function an ideal model of style transfer should realize is discussed. Compared with such a model, potential problems and prospects of different methods to achieve style transfer are listed. A couple of solutions to these drawbacks are given in the end.

In order to detect the objects on the water surface, a neural style learning algorithm is proposed in this paper. The algorithm uses the Gram matrix of a pre-trained convolutional neural network to represent the style of the texture in the image, which is originally used for image style transfer. The objects on the water surface can be easily distinguished by the difference in their styles of the image texture. The algorithm is tested on the dataset of the Airbus Ship Detection Challenge on Kaggle. Compared to the other water surface object detection algorithms, the proposed algorithm has a good precision of 0.925 with recall equals to 0.86.

Deep learning (DL) models have been shown to be vulnerable to adversarial attacks. DL model security against adversarial attacks is critical to using DL-trained models in forward deployed systems, e.g. facial recognition, document characterization, or object detection. We provide results and lessons learned applying a moving target defense (MTD) strategy against iterative, gradient-based adversarial attacks. Our strategy involves (1) training a diverse ensemble of DL models, (2) applying randomized affine input transformations to inputs, and (3) randomizing output decisions. We report a primary lesson that this strategy is ineffective against a white-box adversary, which could completely circumvent output randomization using a deterministic surrogate. We reveal how our ensemble models lacked the diversity necessary for effective MTD. We also evaluate our MTD strategy against a black-box adversary employing an ensemble surrogate model. We conclude that an MTD strategy against black-box adversarial attacks crucially depends on lack of transferability between models.

Mathematical morphology has been explored in deep learning architectures, as a substitute to convolution, for problems like pattern recognition and object detection. One major advantage of using morphology in deep learning is the utility of morphological erosion and dilation. Specifically, these operations naturally embody interpretability due to their underlying connections to the analysis of geometric structures. While the use of these operations results in explainable learned filters, morphological deep learning lacks attribution modeling, i.e., a paradigm to specify what areas of the original observed image are important. Furthermore, convolution-based deep learning has achieved attribution modeling through a variety of neural eXplainable Artificial Intelligence (XAI) paradigms (e.g., saliency maps, integrated gradients, guided backpropagation, and gradient class activation mapping). Thus, a problem for morphology-based deep learning is that these XAI methods do not have a morphological interpretation due to the differences in the underlying mathematics. Herein, we extend the neural XAI paradigm of saliency maps to morphological deep learning, and by doing, so provide an example of morphological attribution modeling. Furthermore, our qualitative results highlight some advantages of using morphological attribution modeling.