Deep Convolutional Neural Networks ((CNN)s) have attained remarkable success in numerous visual recognition tasks. There are two challenges when adopting (CNN)s in real-world applications: a) Existing (CNN)s are computationally expensive and memory intensive, impeding their use in edge computing; b) there is no standard methodology for designing the (CNN) architecture for the intended problem. Network pruning/compression has emerged as a research direction to address the first challenge, and it has proven to moderate (CNN) computational load successfully. For the second challenge, various evolutionary algorithms have been proposed thus far. The algorithm proposed in this paper can be viewed as a solution to both challenges. Instead of using constant predefined criteria to evaluate the filters of (CNN) layers, the proposed algorithm establishes evaluation criteria in online manner during Network training based on the combination of each filter’s profit in its layer and the next layer. In addition, the novel method suggested that it inserts new filters into the (CNN) layers. The proposed algorithm is not simply a pruning strategy but determines the optimal number of filters. Training on multiple (CNN) architectures allows us to demonstrate the efficacy of our approach empirically. Compared to current pruning algorithms, our algorithm yields a Network with a remarkable prune ratio and accuracy. Despite the relatively high computational cost of an epoch in the proposed algorithm in pruning, altogether it achieves the resultant Network faster than other algorithms.

Although, speech recognition systems are widely used and their accuracies are continuously increased, there is a considerable performance gap between their accuracies and human recognition ability. This is partially due to high speaker variations in speech signal. Deep Neural Networks are among the best tools for acoustic modeling. Recently, using hybrid deep Neural Network and hidden Markov model (HMM) leads to considerable performance achievement in speech recognition problem because deep Networks model complex correlations between features. The main aim of this paper is to achieve a better acoustic modeling by changing the structure of deep Convolutional Neural Network ((CNN)) in order to adapt speaking variations. In this way, existing models and corresponding inference task have been improved and extended. Here, we propose adaptive windows Convolutional Neural Network (AW(CNN)) to analyze joint temporal-spectral features variation. AW(CNN) changes the structure of (CNN) and estimates the probabilities of HMM states. We propose adaptive windows Convolutional Neural Network in order to make the model more robust against the speech signal variations for both single speaker and among various speakers. This model can better model speech signals. The AW(CNN) method applies to the speech spectrogram and models time-frequency varieties. This Network handles speaker feature variations, speech signal varieties, and variations in phone duration. The obtained results and analysis on FARSDAT and TIMIT datasets show that, for phone recognition task, the proposed structure achieves 1. 2%, 1. 1% absolute error reduction with respect to (CNN) models respectively, which is a considerable improvement in this problem. Based on the results obtained by the conducted experiments, we conclude that the use of speaker information is very beneficial for recognition accuracy.

Speaker recognition is a process of recognizing persons based on their voice which is widely used in many applications. Although many researches have been performed in this domain, there are some challenges that have not been addressed yet. In this research, Neutrosophic (NS) theory and Convolutional Neural Networks ((CNN)) are used to improve the accuracy of speaker recognition systems. To do this, at first, the spectrogram of the signal is created from the speech signal and then transferred to the NS domain. In the next step, the alpha correction operator is applied repeatedly until reaching constant entropy in subsequent iterations. Finally, a Convolutional Neural Networks architecture is proposed to classify spectrograms in the NS domain. Two datasets TIMIT and Aurora2 are used to evaluate the effectiveness of the proposed method. The precision of the proposed method on two datasets TIMIT and Aurora2 are 93.79% and 95.24%, respectively, demonstrating that the proposed model outperforms competitive models.

Unusual behavior detection is critically important for visual surveillance. It is also a challenging research topic in computer vision. Although much effort has been devoted to tackle this problem, such detection task in a realistic and uncontrolled environment is still far from mature. The major difficulty lies in the ambiguous characteristic in differentiating normal and abnormal behaviors, whose definitions often vary according to the context of video's history. In this paper we propose a framework for detecting and locating abnormal activities in video sequences. The key aspect of our method is the pairing of the 2D and 3D spatial-temporal Convolutional Neural Networks ((CNN)) for anomaly detection in contiguous video frames. The Features from Accelerated Segment Test (FAST) detector has been used In order to increase the reliability in identifying the interest locations in entry frames of Convolutional Neural Network model. These feature extracted only from volumes of moving pixels that reduce the computational costs. The architecture of (CNN) model allows us to extract spatial-temporal features that contain complicated motion features. We test our framework on popular benchmark dataset containing various human abnormal activities and situations. Evaluation results show that our method outperforms most of other methods and achieves a very competitive detection performance compared to state-of-the-art methods.