Today, public areas, such as airports, hospitals, city centers are monitored by surveillance systems. The widespread use of surveillance systems reduces security concerns while creating an amount of video data that cannot be examined by people in real-time. Therefore, the concept of automatic understanding of video activities has raised the standards of security camera systems. In this paper, we propose a framework (OF-ConvAE-LSTM) to detect anomalies using Convolutional Autoencoder and Convolutional Long Short-Term Memory in an unsupervised manner. Besides the deep learning model, the feature extraction stage based on dense optical flow is applied in the framework to obtain the velocity and direction information of foreground objects. The experiments were carried out on three popular public datasets consisting of Avenue, UCSD Ped1, and UCSD Peds2. The experimental results have shown that the proposed framework models the complex distribution of the pattern of regular motion changes with high accuracy. Besides, this method was observed to outperform state-of-the-art approaches based on unsupervised and semi-supervised deep learning models.INDEX TERMS Abnormal event detection, convolutional autoencoder, long short-term memory, optical flow.

Light emitting diode (LED) based visible light positioning networks can provide accurate location information in environments where the global positioning system (GPS) suffers from severe signal degradation and/or cannot achieve high precision, such as indoor scenarios. In this paper, we propose to employ cooperative localization for hybrid infrared/visible light networks that involve multiple LED transmitters having known locations (e.g., on the ceiling) and visible light communication (VLC) units equipped with both LEDs and photodetectors (PDs) for the purpose of cooperation. In the considered scenario, downlink transmissions from LEDs on the ceiling to VLC units occur via visible light signals, while the infrared spectrum is utilized for deviceto-device communications among VLC units. First, we derive the Cramér-Rao lower bound and the maximum likelihood estimator (MLE) for the localization of VLC units in the proposed cooperative scenario. To tackle the nonconvex structure of the MLE, we adopt a set-theoretic approach by formulating the problem of cooperative localization as a quasiconvex feasibility problem, where the aim is to find a point inside the intersection of convex constraint sets constructed as the sublevel sets of quasiconvex functions resulting from the Lambertian formula. Next, we devise two feasibility-seeking algorithms based on iterative gradient projections to solve the feasibility problem. Both algorithms are amenable to distributed implementation, thereby avoiding high-complexity centralized approaches. Capitalizing on the concept of quasi-Fejér convergent sequences, we carry out a formal convergence analysis to prove that the proposed algorithms converge to a solution of the feasibility problem in the consistent case. Numerical examples illustrate the improvements in localization performance achieved via cooperation among VLC units and evidence the convergence of the proposed algorithms to true VLC unit locations in both the consistent and inconsistent cases.

In this paper, cooperative localization is proposed for visible light systems. The effects of cooperation on the localization accuracy of visible light positioning systems are illustrated based on a Cramér-Rao lower bound expression. The obtained expression is generic for any three-dimensional configuration and covers all possible cooperation scenarios via definitions of connectivity sets. Numerical results are presented to investigate significance of cooperation in various scenarios.

Accurate and efficient localization of the optic disk (OD) in retinal images is an essential process for the diagnosis of retinal diseases, such as diabetic retinopathy, papilledema, and glaucoma, in automatic retinal analysis systems. This paper presents an effective and robust framework for automatic detection of the OD. The framework begins with the process of elimination of the pixels below the average brightness level of the retinal images. Next, a method based on the modified robust rank order was used for edge detection. Finally, the circular Hough transform (CHT) was performed on the obtained retinal images for OD localization. Three public datasets were used to evaluate the performance of the proposed method. The optic disks were successfully located with the success rates of 100%, 96.92%, and 98.88% for the DRIVE, DIARETDB0, and DIARETDB1 datasets, respectively.

No abstract