Ying Tian^{1, *}, Libing Wang¹, Hexin Gu², Lin Fan³

CMC-Computers, Materials & Continua, Vol.65, No.3, pp. 2397-2412, 2020, DOI:10.32604/cmc.2020.011386 - 16 September 2020

Abstract The application of deep learning in the field of object detection has experienced much progress. However, due to the domain shift problem, applying an off-the-shelf detector to another domain leads to a significant performance drop. A large number of ground truth labels are required when using another domain to train models, demanding a large amount of human and financial resources. In order to avoid excessive resource requirements and performance drop caused by domain shift, this paper proposes a new domain adaptive approach to cross-domain vehicle detection. Our approach improves the cross-domain vehicle detection model from More >

Jieren Cheng¹, Jun Li^{2, *}, Naixue Xiong³, Meizhu Chen², Hao Guo², Xinzhi Yao²

CMC-Computers, Materials & Continua, Vol.65, No.3, pp. 2247-2262, 2020, DOI:10.32604/cmc.2020.011668 - 16 September 2020

Abstract Nowadays, as lightweight mobile clients become more powerful and widely used, more and more information is stored on lightweight mobile clients, user sensitive data privacy protection has become an urgent concern and problem to be solved. There has been a corresponding rise of security solutions proposed by researchers, however, the current security mechanisms on lightweight mobile clients are proven to be fragile. Due to the fact that this research field is immature and still unexplored in-depth, with this paper, we aim to provide a structured and comprehensive study on privacy protection using trusted execution environment… More >

Shaozhang Niu^{1, *}, Xiangxiang Li¹, Maosen Wang¹, Yueying Li²

CMC-Computers, Materials & Continua, Vol.65, No.3, pp. 2233-2245, 2020, DOI:10.32604/cmc.2020.09471 - 16 September 2020

Abstract In recent years, images have played a more and more important role in our daily life and social communication. To some extent, the textual information contained in the pictures is an important factor in understanding the content of the scenes themselves. The more accurate the text detection of the natural scenes is, the more accurate our semantic understanding of the images will be. Thus, scene text detection has also become the hot spot in the domain of computer vision. In this paper, we have presented a modified text detection network which is based on further… More >

Qiqiang Chen^{1, *}, Xinxin Gan², Wei Huang¹, Jingjing Feng¹, H. Shim³

CMC-Computers, Materials & Continua, Vol.65, No.3, pp. 2201-2215, 2020, DOI:10.32604/cmc.2020.011191 - 16 September 2020

Abstract Automatic road damage detection using image processing is an important aspect of road maintenance. It is also a challenging problem due to the inhomogeneity of road damage and complicated background in the road images. In recent years, deep convolutional neural network based methods have been used to address the challenges of road damage detection and classification. In this paper, we propose a new approach to address those challenges. This approach uses densely connected convolution networks as the backbone of the Mask R-CNN to effectively extract image feature, a feature pyramid network for combining multiple scales More >

Huifang Qian¹, Xuan Zhou^{1, *}, Mengmeng Zheng¹

CMC-Computers, Materials & Continua, Vol.65, No.3, pp. 2153-2167, 2020, DOI:10.32604/cmc.2020.011843 - 16 September 2020

Abstract The core technology in an intelligent video surveillance system is that detecting and recognizing abnormal behaviors timely and accurately. The key breakthrough point in recognizing abnormal behaviors is how to obtain the effective features of the picture, so as to solve the problem of recognizing them. In response to this difficulty, this paper introduces an adjustable jump link coefficients model based on the residual network. The effective coefficients for each layer of the network can be set after using this model to further improving the recognition accuracy of abnormal behavior. A convolution kernel of 1×1… More >

Hua Yan^1,2,3,*, Bo Song^1,3, Mansheng Wang⁴

Structural Durability & Health Monitoring, Vol.14, No.3, pp. 265-282, 2020, DOI:10.32604/sdhm.2020.06403 - 14 September 2020

Abstract Analyze and compare the basic principles of ultrasonic detection of voids in concrete, choose ZBL-U520/510 non-metallic ultrasonic detector, and use the opposite detection method to test the void size in the joints of prefabricated concrete structures. The results show that: ultrasonic method by testing the waveform, sound, and speed of sound analysis can effectively determine the position of the defect, and through the conversion formula can estimate the void size. Ultrasonic parameters are used to distinguish the internal defects of Assembly concrete. Sometimes there are different results with different parameters. It is difficult for engineers… More >

Alan Alonso-Rivers¹, Rolando Salgado-Estrada^2,*

Structural Durability & Health Monitoring, Vol.14, No.3, pp. 187-204, 2020, DOI:10.32604/sdhm.2020.010743 - 14 September 2020

Abstract A dynamic simulation method for cracked structures is implemented to determine their dynamic response with the purpose of evaluating their structural behavior. The procedure makes possible the simulation of three-dimensional cracked structures. The excitation force is randomly generated to simulate wind gusts. It is assumed the structure remains in the elastic range, which allows for each mode that contributes to its dynamic response to be decoupled. The results indicate that the presence of damage causes changes in the modals parameters of the structure as accurate as other similar methods proposed for simpler structures. Therefore, it More >

Ruilong Chen¹, Guangfu Zeng¹, Ke Wang², Lei Luo^1,*, Zhiping Cai¹

Journal on Internet of Things, Vol.2, No.2, pp. 55-64, 2020, DOI:10.32604/jiot.2020.09814 - 14 September 2020

Abstract Smoking is the main reason for fire disaster and pollution in petrol station, construction site and warehouse. Existing solutions based on wearable devices and smoking sensors were costly and hard to obtain evidence of smoking in unmanned scenarios. With the developments of closed circuit television (CCTV) system, vision-based methods for object detection, mostly driven by deep learning techniques, were introduced recently. However, the massive GPU computing hardware required by the deep learning algorithm made these methods hard to be deployed. This paper aims at solving the smoking detection problem on edge and proposes the solution More >

Xing Deng^1,2, Haijian Shao^1,2,*

Energy Engineering, Vol.117, No.5, pp. 279-287, 2020, DOI:10.32604/EE.2020.011619 - 07 September 2020

Abstract Recurrent neural networks (RNNs) as one of the representative deep learning methods, has restricted its generalization ability because of its indigestion hidden-layer information presentation. In order to properly handle of hidden-layer information, directly reduce the risk of over-fitting caused by too many neuron nodes, as well as realize the goal of streamlining the number of hidden layer neurons, and then improve the generalization ability of RNNs, the hidden-layer information of RNNs is precisely analyzed by using the unsupervised clustering methods, such as Kmeans, Kmeans++ and Iterative self-organizing data analysis (Isodata), to divide the similarity of More >

Yaping Mao¹, Kaiyong Li², Duolu Mao^2,*

Journal of New Media, Vol.2, No.3, pp. 131-135, 2020, DOI:10.32604/jnm.2020.012815 - 04 September 2020

Abstract Based on the analysis of the advantages and disadvantages of GPS positioning system in practical application, this paper proposes the combination of wireless network positioning technology and GPS positioning system to overcome the low accuracy of GPS positioning system in the case of occlusion. This paper introduces in detail the principle of the application of wireless network positioning technology based on GPS positioning system in geographic information measurement, and illustrates its practical application in production by taking coal mine positioning as an example. More >