Solving 3D Coverage Problem using Genetic Algorithms in Wireless Camera-Based Sensor Network Modelling

<p>Article publication in IEEE Acess uploaded here to follow EU rules mandated for projects under Horizon 2020.</p>

Efficient deployment of cameras optimizes the costs and benefits in a multicamera surveillance network. Most camera placement algorithms are developed for surveillance regions with predefined camera locations, where optimal orientation for every given camera location is determined to maximize coverage. The existing approaches to optimizing the number of cameras, locations, and orientations focus on specific scenarios and are neither generic nor scalable across diverse surveillance scenarios. This article proposes a generic and scalable reward penalty score (RPS) algorithm for optimal camera placement in diverse indoor and outdoor scenarios to efficiently attain a specified coverage requirement. The camera placement decision is driven by maximizing total coverage over the surveillance region with a minimum number of cameras at optimal locations and orientations. Based on the RPS algorithm, we extend the existing Greedy grid voting (GGV) algorithm that was limited to optimizing camera orientations for predefined locations. The extended GGV (EGGV) algorithm is broader in scope and optimizes the number of cameras, their locations, and orientations. Overall, the RPS algorithm produces superior coverage results over indoor and outdoor scenarios compared to the existing camera placement algorithms on map images of diverse surveillance scenarios ranging from small to large.

The efficient management (placement and orientation) of security cameras within a floor plan is a well-known and difficult problem that has gained attention recently. The objective is to locate the minimum number of cameras in the space to ensure all walls are within the view of at least one camera. Heuristic-based approaches have been developed for this NP-hard problem, unfortunately, most are only applicable to static situations. In modern applications, surveillance management must be resilient, and adapt if the environment changes. This paper introduces evolutionary-based approaches for active surveillance camera management. Using an evolutionary-based approach, a surveillance configuration (camera locations and orientations) is encoded as a chromosome and evolutionary processes are applied to identify better solutions over successive generations. The approach has the ability to identify efficient surveillance configurations (minimum number of cameras with maximum coverage), however, another advantage is the ability to adapt if the environment unexpectedly changes. Simulation results demonstrate this type of approach can manage surveillance cameras under dynamic conditions such as camera loss and the introduction obstacles better than traditional search methods.

This paper addresses the camera placement problem for smart cities in 3-D space and proposes a heuristic algorithm that maximizes the weighted coverage rate while satisfying the budget constraint. We first discuss about the quantization of surveillance target and camera locations into discrete grid points and the setting of the weights of target grid points. We then present the visibility analysis with field of view in consideration of occlusions and different camera specifications. Based on these characteristics and constraints, we formulate the camera placement problem and propose a new heuristic algorithm called collaboration-based local search algorithm, which incorporates the local search into collaborative allocation. We evaluate the performance of the proposed algorithm in comparison with the greedy algorithm, binary genetic algorithm, and binary particle swarm optimization through simulation experiments with small and large problem sets. The simulation results show that the proposed algorithm outperforms the three existing algorithms in terms of the average weighted coverage rate and computation time.

A new active-vision system for 3-D object recognition in robotic assembly workcells is being developed. In this system, image analysis is simplified by premarking the objects with artificial markers and utilizing a mobile camera to acquire views of the objects prespecified by markers. Hence, the objects are represented by a small set of 2-D perspective projections. In this context, an important problem to be addressed is the determination of the number of markers and their locations on the objects, and of the placement of the mobile camera to guarantee the detection of at least one marker. The authors present a solution to the optimal arrangement of markers and camera location(s), given the objects' solid models and a set of constraints. >

Introduction. To develop reliable solutions, robotic systems use several channels for contactless control of moving objects. It is especially active in contactless robotic systems for evaluating the quality of agricultural products. Determination of informative parameters of fruits (shape, size, surface texture and color) is a characteristic feature of the complexes, which are used to make final estimation fruit’s quality. Purpose. The research of the enhancement robotic optoelectronic channels, that allow increasing of the reliability of decision-making during separation in the flow of complex shape fruits (pitaya, pineapple and custard apple) by combining information about their shape, size, surface texture and color from several images, which are obtained for various fields of observation with illumination of the optical pulsed sources. Methods. To solve this problem, the authors propose research on digital multi-angle color images processing of fruits from several video cameras with their synchronous multidirectional pulse sources of illumination. The number, spatial mutual location of video cameras and sources, as well as the time procedures for their interaction allow improving the accuracy of determining the parameters of shape, size, surface texture and color, which ultimately will increase the reliability of decision-making about the quality of the fruit. Practical relevance. Computer simulation of the spatial placement of video cameras (up to 5 pieces) and their calibration methods with impulse sources of illumination can increase the accuracy of determining geometric shape parameters (curvature, center of gravity, convexity, roundness coefficient and eccentricity) by 8%. Experimental results on an installation with video cameras and eight impulse sources of illumination confirmed the increase image contrast by 2 times.

Marker-based 3D motion capture systems, widely used for gait analysis, are accurate but have disadvantages such as cost and accessibility. Whereas markerless pose estimation has emerged as a convenient and cost-effective alternative for gait analysis, challenges remain in achieving optimal accuracy. Given the limited research on the effects of camera location and orientation on data collection accuracy, this study investigates how camera placement affects gait assessment accuracy utilizing five smartphones. This study aimed to explore the differences in data collection accuracy between marker-based systems and pose estimation, as well as to assess the impact of camera location and orientation on accuracy in pose estimation. The results showed that the differences in joint angles between pose estimation and marker-based systems are below 5°, an acceptable level for gait analysis, with a strong correlation between the two datasets supporting the effectiveness of pose estimation in gait analysis. In addition, hip and knee angles were accurately measured at the front diagonal of the subject and ankle angle at the lateral side. This research highlights the significance of careful camera placement for reliable gait analysis using pose estimation, serving as a concise reference to guide future efforts in enhancing the quantitative accuracy of gait analysis.

Tiny block-size coding for energy-efficient image compression and communication in wireless camera sensor networks

Wireless camera sensor networks have to balance the conflicting challenges imposed by the detection performance, latency and lifetime requirements in surveillance applications. While previous studies for camera sensor networks have addressed these issues separately, they have not quantified the trade-offs between these requirements. In this paper, we discuss the design and implementation of SCOPES, a distributed Smart Camera Object Position Estimation System that balances the trade-offs associated with camera sensor networks. The main contribution of the paper is the extensive evaluation of parameters affecting the performance of the system through analysis, simulation and experimentation in real-life conditions. Our results demonstrates the effectiveness of SCOPES, which achieves detection probabilities ranging from 84% to 98% and detection latencies from 10 seconds to 18 seconds. Moreover, by using coordination schemes, the detection performance of SCOPES was improved with increased system lifetime. SCOPES highlights that intelligent system design can compensate for resource-constrained hardware and computationally simple data processing algorithms.

Computer‐Aided Optimization of Surveillance Cameras Placement on Construction Sites

Wireless sensor and actuator networks have been extensively deployed for enhancing industrial control processes and supply-chains, and many forms of surveillance and environmental monitoring. The availability of low-cost mobile robots equipped with a variety of sensors in addition to communication and computational capabilities makes them particularly promising in target coverage tasks for ad hoc surveillance, where quick, low-cost or non-lasting visual sensing solutions are required, e.g. in border protection and disaster recovery. In this paper, we consider the problem of low complexity placement and orientation of mobile cameras to cover arbitrary targets. We tackle this problem by clustering proximal targets, while calculating/estimating the camera location/direction for each cluster separately through our cover-set coverage method. Our proposed solutions provide extremely computationally efficient heuristics with only a small increase in number of cameras used, and a small decrease in number of covered targets. 2014 IEEE.

Entropy encoder for low-power low-resources high-quality CFA image compression

In camera sensor network research, physical camera sensor network platforms with low power and low cost are needed for testing and validating algorithms. It is also necessary that the camera nodes be calibrated precisely. In this work, we first develop and evaluate a low power and low cost wireless camera sensor network platform. The camera sensor nodes in this platform transmit a grayscale image over a wireless channel to a master control station. Then we propose a simple, light-weight algorithm to perform distributed calibration of the camera sensor nodes. The camera sensor nodes use their imaging abilities in collaboration with a cooperative moving target to determine their own positions and orientations. The proposed algorithm requires simple arithmetic calculations and hence it can be realized on low power processors. This calibration algorithm has been implemented and evaluated on the developed physical camera sensor network platform.

Design and performance analysis of a zonal DCT-based image encoder for Wireless Camera Sensor Networks

Due to technological advances and the popularity of camera sensors, it is now straightforward for users to capture and share videos. A large number of geo-tagged photos and videos have been accumulating continuously on the web, posing a challenging problem for mining this type of media data. In one application scenario, users might desire to know what the Points of Interest (POI) are which contain important objects or places in a video. Existing solutions attempt to examine the content of the videos and recognize objects and events. This is typically time-consuming and computationally expensive and the results can be uneven. Therefore these methods face challenges when applied to large video repositories. We propose a novel technique that leverages sensor-generated meta-data (camera locations and viewing directions) which are automatically acquired as continuous streams together with the video frames. Existing smartphones can easily accommodate such integrated recording tasks. By considering a collective set of videos and leveraging the acquired auxiliary meta-data, our approach is able to detect interesting regions and objects (POIs) and their distances from the camera positions in a fully automated way. Because of its computational efficiency, the proposed method scales well and our experiments show very promising results.