Environment
A. Sezavar; H. Farsi; S. Mohamadzadeh
Abstract
Person re-identification (re-id) is one of the most critical and challenging topics in image processing and artificial intelligence. In general, person re-identification means that a person seen in the field of view of one camera can be found and tracked by other non-overlapped cameras. Low-resolution ...
Read More
Person re-identification (re-id) is one of the most critical and challenging topics in image processing and artificial intelligence. In general, person re-identification means that a person seen in the field of view of one camera can be found and tracked by other non-overlapped cameras. Low-resolution frames, high occlusion in crowded scene, and few samples for training supervised models make re-id challenging. This paper proposes a new model for person re-identification to overcome the noisy frames and extract robust features from each frame. To this end, a noise-aware system is implemented by training an auto-encoder on artificially damaged frames to overcome noise and occlusion. A model for person re-identification is implemented based on deep convolutional neural networks. Experimental results on two actual databases, CUHK01 and CUHK03, demonstrate that the proposed method performs better than state-of-the-art methods.
Environment
Z. Dorrani; H. Farsi; S. Mohamadzadeh
Abstract
In traffic monitoring for video analysis systems, vehicle shadows have a negative effect on their performance. Shadow detection and removal are essential steps in accurate vehicle detection. In this paper, a new method is proposed for shadow detection using a novel convolution neural network architecture. ...
Read More
In traffic monitoring for video analysis systems, vehicle shadows have a negative effect on their performance. Shadow detection and removal are essential steps in accurate vehicle detection. In this paper, a new method is proposed for shadow detection using a novel convolution neural network architecture. In the proposed method, the edges of the image are first extracted. Edge extraction reduces calculation, and accelerates the execution of the method. The background of the frame is then removed and the main features are extracted using the ResUNet-a architecture. This architecture consists of two parts: the encoder and the decoder, which detect the shadow at the decoder output and then remove it. Deep learning is used to detect shadows, which increases the accuracy of the analysis. The ResUNet-a architecture can learn complex, hierarchical, and appropriate features from the image for accurate feature detection and discarding the irrelevant shadow, thereby outperforming conventional filters.The results show that the proposed method provides better performance on NJDOT traffic video, highway-1, and highway-3 datasets than popular shadow removal methods. Also, the method improves the evaluation criteria such as F-measure and runtime. The F-measure is 94 and 93% for highway-1 and highway-3, respectively.
A. Ataee; J. Kazemitabar; M. Najafi
Abstract
Due to lack of proper regulations in many areas of the world, consumers are not mandated to waste sorting at the origin of the source. Moreover, human sorting often suffers from human errors and low accuracy. In the intelligent detection system, it is attempted to break down a variety of household wastes ...
Read More
Due to lack of proper regulations in many areas of the world, consumers are not mandated to waste sorting at the origin of the source. Moreover, human sorting often suffers from human errors and low accuracy. In the intelligent detection system, it is attempted to break down a variety of household wastes including plastic bottles, glass, metals, paper bags, compact plastics, paper and disposable containers. In this paper, a real waste image system is investigated using the deep convolutional neural network and a remarkable accuracy of 92.76% was achieved.
Environment
M. Moallem; H. Hassanpour; A. A. Pouyan
Abstract
Smart homes enable many people, especially the elderly and patients, to live alone and maintain their independence and comfort. The realization of this goal depends on monitoring all activities in the house to report any observed anomaly immediately to their relatives or nurses. Anomaly detection in ...
Read More
Smart homes enable many people, especially the elderly and patients, to live alone and maintain their independence and comfort. The realization of this goal depends on monitoring all activities in the house to report any observed anomaly immediately to their relatives or nurses. Anomaly detection in smart homes, just by existing data, is not an easy task. In this work, we train a recurrent network with raw outputs of binary sensors, including motion and door sensors, to predict which sensor will be switched on/off in the next event, and how long this on/off mode will last. Then, using Beam Search, we extend this event into k sequences of consecutive events to determine the possible range of upcoming activities. The error of this prediction i.e., the distance between these possible sequences and the real string of events is evaluated using several innovative methods for measuring the spatio-temporal similarity of the sequences. Modeling this error as a Gaussian distribution allows to assess the likelihood of anomaly scores. The input sequences that are ranked higher than a certain threshold will be considered as abnormal activities. The results of the experiments showed that this method enables the detection of abnormal activities with desirable accuracy.