Search

Innovations on the Internet of Everything (IoE) enabled systems are driving a change in the settings where we interact in smart units, recognized globally as smart city environments. However, intelligent video-surveillance systems are critical to increasing the security of these smart cities. More precisely, in today’s world of smart video surveillance, person re-identification (Re-ID) has gained increased consideration by researchers. Various researchers have designed deep learning-based algorithms for person Re-ID because they have achieved substantial breakthroughs in computer vision problems. In this line of research, we designed an adaptive feature refinement-based deep learning architecture to conduct person Re-ID. In the proposed architecture, the inter-channel and inter-spatial relationship of features between the images of the same individual taken from nonidentical camera viewpoints are focused on learning spatial and channel attention. In addition, the spatial pyramid pooling layer is inserted to extract the multiscale and fixed-dimension feature vectors irrespective of the size of the feature maps. Furthermore, the model’s effectiveness is validated on the CUHK01 and CUHK02 datasets. When compared with existing approaches, the approach presented in this paper achieves encouraging Rank 1 and 5 scores of 24.6% and 54.8%, respectively. [ABSTRACT FROM AUTHOR]

In computer vision, human activity recognition is an active research area for different contexts, such as human–computer interaction, healthcare, military applications, and security surveillance. Activity recognition is performed to recognize the goals and actions of one or more people from a sequence of observations based on the actions and the environmental conditions. Still, there are numbers of challenges and issues, which motivate the development of new activity recognition method to enhance the accuracy under more realistic conditions. This paper proposes an error-based fuzzy dragon deep belief network (error-based DDBN), which is the integration of fuzzy with DDBN classifier, to recognize the human activity from a complex and diverse scenario, for which the keyframes are generated based on the Bhattacharya coefficient from the set of frames of the given video. The key frames from the Bhattacharya are extracted using the scale invariant feature transform, color histogram of the spatio-temporal interest dominant points, and hierarchical skeleton. Finally, the features are fed to the classifier, where the classification is done using the proposed error-based fuzzy DDBN to recognize the person. The experimentation is performed using two datasets, namely KTH and Weizmann for analyzing the performance of the proposed classifier. The experimental results reveal that the proposed classifier performs the activity recognition in a better way by obtaining the maximum accuracy of 1, a sensitivity of 0.99, and the specificity of 0.991. [ABSTRACT FROM AUTHOR]

Recent advances in large‐scale image and video analysis have empowered the potential capabilities of visual surveillance systems. In particular, deep learning‐based approaches bring in substantial benefits in solving certain computer vision problems such as fine‐grained object recognition. Here, the authors mainly concentrate on classification and identification of maritime vessels and land vehicles, which are the key constituents of visual surveillance systems. Employing publicly available data sets for maritime vessels and land vehicles, the authors aim to improve visual recognition. Specifically, the authors focus on five tasks regarding visual recognition; coarse‐grained classification, fine‐grained classification, coarse‐grained retrieval, fine‐grained retrieval, and verification. To increase the performance in these tasks, the authors utilise a multi‐task learning framework and present a novel loss function which simultaneously considers deep feature learning and classification by exploiting the available hierarchical labels of individual samples and the global statistics of distances between the data pairs. The authors observe that the proposed multi‐task learning model improves the fine‐grained recognition performance on MARVEL and Stanford Cars data sets, compared to training of a model targeting a single recognition task.

A variety of visual surveillance systems has been developed to satisfy different purposes. However, traditional visual surveillance systems are inflexible and passive, and thus incur significantly high overhead upon recording video information. Therefore, to improve such systems, researchers have been incorporating some special technologies into traditional visual surveillance systems. Doing so makes the advantages of these adopted technologies able to be attached to these systems such that the systems can benefit much from these technologies. In this study, the authors devise a novel visual surveillance system based on a distributed collaborative mechanism among all monitoring cameras. More specifically, the authors utilise the concept of a well‐known distributed application, i.e. leader election, to implement this mechanism. Such a mechanism helps a visual surveillance system with more than one monitoring camera automatically and efficiently assign the proper number of cameras to track a particular object. Hence, upon a moderate number of objects entering the monitoring area, the monitoring cameras will be appointed to each object almost evenly. [ABSTRACT FROM AUTHOR]

Biometric researchers have recently found considerable applicability of gait recognition in visual surveillance systems. This study proposes a probabilistic framework for gait modelling that is applied to gait recognition. The basic idea of this framework is to consider the silhouette shape as a multivariate random variable and model it in a full probabilistic framework. The Bernoulli mixture model is employed to model silhouette distribution and recursive algorithms are provided for silhouette image and sequence classification. Finally, the proposed probabilistic method is applied to benchmark databases and its validity is demonstrated through experiments.

Detecting objects of interest and obtaining their clear visual appearances are critical requirements for visual surveillance systems. In this paper we propose a novel algorithm to detect foreground objects from video sequences with fog and then enhance their visibilities. First, we propose a novel metric to measure the image fog property to decide whether the image scene is obscured by fog or not. Second, if there is heavy fog in the scene, a novel approach for object detection based on an atmospheric scattering model is proposed. This novel approach can be used to detect not only newly entering objects but also sojourned objects. Once the foreground object is detected, we enhance its visibility only to avoid processing the whole image. Our proposed algorithm is tested with some surveillance video under different fog conditions. Experimental results show that the proposed approach is efficient and efficient for foreground object detection and visibility enhancement under fog weather conditions. [ABSTRACT FROM AUTHOR]

Abstract: In this work we propose a general top–down feedback scheme between adjacent description levels to interpret video sequences. This scheme distinguishes two types of feedback: repair-oriented feedback and focus-oriented feedback. With the first it is possible to improve the system''s performance and produce more reliable and consistent information, and with the second it is possible to adjust the computational load to match the aims. Finally, the general feedback scheme is used in different examples for a visual surveillance application which improved the final result of each description level by using the information in the higher adjacent level. [Copyright &y& Elsevier]

Abstract: A key problem in visual surveillance systems (VSS) is to find an effective procedure for linking the geometric descriptions of a scene at the object level with the corresponding descriptions of the agents intervening in this scene at the activity level. In this work, we explore a constructivist approach based on using the usual Artificial Intelligence (AI) techniques and methods to establish correspondences between the entities and relations of the ontologies in these two levels. The proposal is exemplified using a real interior scenario that uses images from just one fixed camera and where the purpose of the surveillance is to do a preventive diagnosis of the activity of abandoning a potentially dangerous object in a sensitive area. The work stresses: (1) anchoring the object-level labels in the result of analytical processes on blobs, (2) specifying contextual knowledge that has to be injected to link the activities, as described by a human surveillance expert, with the objects, as they are labelled by the same expert from geometric descriptions. The work is set within the context of the 50th anniversary of AI and the leading theories on human visual perception. [Copyright &y& Elsevier]