Design of an Active Laser Mini-Camera Detection System Using CNN

The growing popularity of the mini-camera is posing a serious threat to privacy and personal security. Disguised as common tools in rooms, these devices can become undetectable. Moreover, conventional active laser detection systems often fail to recognize them owing to their small lens size, weak reflectivity, and the influence of interference targets. In this paper, a method for building a laser active detection system for mini-cameras is proposed. Using a monostatic optical system and a deep learning classification algorithm, this anti-camera system can detect mini-cameras accurately in real time. This article describes the system components including its optical design, core components and image processing algorithm. The capability of the system for detecting mini-cameras and identifying interference is also experimentally demonstrated. This work successfully overcomes the limit of mini-camera detection using deep learning methods in active laser detection systems.

[1]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Mali Gong,et al.  Periodicity analysis on cat-eye reflected beam profiles of optical detectors , 2017 .

[3]  Ching Y. Suen,et al.  Scene Classification Using Hierarchical Wasserstein CNN , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Chang-Tsun Li,et al.  Social Network Identification Through Image Classification With CNN , 2019, IEEE Access.

[7]  Hui Huang,et al.  Multi-Spectral RGB-NIR Image Classification Using Double-Channel CNN , 2019, IEEE Access.

[8]  Jianjun Lei,et al.  Joint spatial-spectral hyperspectral image classification based on convolutional neural network , 2020, Pattern Recognit. Lett..

[9]  Xiangyu Zhang,et al.  ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[10]  Jui-che Tsai,et al.  Design, Fabrication, and Characterization of Tunable Cat's Eye Retroreflector Arrays as Optical Identification Tags , 2014, Journal of Lightwave Technology.

[11]  Li Li,et al.  Recognizing cat-eye targets with dual criterions of shape and modulation frequency , 2011 .

[12]  Mali Gong,et al.  Cat-eye effect reflected beam profiles of an optical system with sensor array. , 2016, Applied optics.

[13]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[14]  Bo Liu,et al.  Compressive sensing method for recognizing cat-eye effect targets. , 2013, Applied optics.

[15]  Li Li,et al.  Fast cat-eye effect target recognition based on saliency extraction , 2015 .

[16]  Lixin Zheng,et al.  Combining Convolutional Neural Network With Recursive Neural Network for Blood Cell Image Classification , 2018, IEEE Access.