Entity

Time filter

Source Type


Guo S.,Shenzhen Institute of Information Technology | Guo S.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | Chen S.-F.,Shenzhen Institute of Information Technology
Proceedings - International Conference on Machine Learning and Cybernetics | Year: 2012

In this paper, a new method is proposed to estimate crowd density based on binocular vision: according to the average parallax of the object in the two camera plane, first we calculate the correction parameters of monitoring location, then we fit the correction function based on the result of the correction parameters of the same target objects in different locations. By taking foreground pixels, foreground edge pixels of image as features of crowd density estimation, we could perform parameter correction. Experiments show that by comparing with existing methods, this method is effective in reducing the projective deformity and can improve the effectiveness of the features greatly. So it could enhance the accuracy of the estimation of population density. © 2012 IEEE. Source


Hu T.,Shenzhen Institute of Information Technology | Hu T.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | Liu Z.,Shenzhen Institute of Information Technology | Cui J.,Shenzhen Institute of Information Technology | And 2 more authors.
Proceedings of SPIE - The International Society for Optical Engineering | Year: 2013

This paper introduces a coaxial visible and infrared dual-band imager, which utilizes the visible and infrared detection technology, uptakes the scenery radiation of different wavelengths or optical energy reflected, adopts transmission and secondary reflection theory to design the coaxial optical path, and realizes the dual-band imaging of same scene. The imager can acquire the registered visible and infrared images, and effectively solves the problem of registration for visible and infrared images in different-source image fusion. © 2013 SPIE. Source


Lu X.,Shenzhen Institute of Information Technology | Lu X.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | Xu J.,Algorithm | Lin G.,Shenzhen Institute of Information Technology | And 2 more authors.
International Journal of Digital Content Technology and its Applications | Year: 2012

Channel estimation is an important issue in any OFDM system for demodulation and decoding. In this paper, the channel estimation based on LS(Least-squares Estimation) in IEEE Std.802.16d/e is studied to resolve the error floor for SUI-5 channel. Since the pilot allocation is solid, we can't modify it exhaustively. So we combine the preamble and pilot carrier to resolve the problem. Here we introduce preamble time-domain LS channel estimation algorithm based on short preamble and pilot for consult. Theoretical analysis and computer simulations show that if we select channel estimation algorithm properly, combing preamble and pilot can improve the performance. We also conclude that pilot only scheme exits error floor when the coherent bandwidth of channel is narrow, and preamble only scheme can decrease the error floor drastically when the coherent bandwidth of channel is narrow, however it can't trace the fast fading channel. Source


Lu X.,Shenzhen Institute of Information Technology | Lu X.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | Xu J.,Algorithm | Lin G.,Shenzhen Institute of Information Technology | And 2 more authors.
Advances in Information Sciences and Service Sciences | Year: 2011

Channel estimation is an important issue in any OFDMA system for demodulation and decoding. The channel estimation based on Comb type pilot arrangement in IEEE Std.802.16d/e is studied through different algorithms for both estimating channel at pilot frequencies and interpolating the channel. The estimation of channel at pilot frequencies is based on LS while the channel interpolation is done using linear interpolation, second-order interpolation and third-order interpolation. We have compared the performances of all schemes by measuring bit error rate with QPSK and BPSK as modulation schemes, and multipath Rayleigh fading channels as channel models. Theoretical analysis and computer simulations show that the third-order interpolation schemes can yield slightly better performance without considering the complexity. Source


Hu T.,Shenzhen Institute of Information Technology | Hu T.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | Ren X.-Y.,Shenzhen Institute of Information Technology | Ren X.-Y.,Shenzhen Key Laboratory of Visual Media Processing and Transmission | And 2 more authors.
Guangdian Gongcheng/Opto-Electronic Engineering | Year: 2013

The high-resolution PCB bare board images in online high-precision AOI detection system are up to 60000×60000 or more. In order to improve processing speed, image contours extraction and DP polyline simplification are used to reduce the amount of processing data. Distance measure of point to segment is one of determinants which affect the efficiency of DP algorithm. A new method to compute distance is proposed. Firstly, a rotating coordinate is established based on the two endpoints of curve, in which the new coordinate value is computed for each point and used to divide the points into three zones and calculate distance, and Manhattan distance is adopted in zone I and III, perpendicular distance in zone II. Compared with Dan Sunday's method, the proposed method takes full advantage of the computation result of the previous point, and the way to divide the points is more concise and efficient, the distance metric calculation amount for points in zone I and III basically keeps, but the amount for points in zone II which own highest proportion reduces significantly. Experimental results show that the improved distance measure method can improve the efficiency of DP polyline simplification algorithm for high resolution PCB bare board image contours. Source

Discover hidden collaborations