中国科学院机构知识库网格系统: 检索

浏览/检索结果: 共6条，第1-6条

帮助

条数/页：排序方式：
	Structured Sparse Coding With the Group Log-regularizer for Key Frame Extraction 期刊论文 OAI收割 IEEE/CAA Journal of Automatica Sinica, 2022, 卷号: 9, 期号: 10, 页码: 1818-1830 作者: Zhenni Li; Yujie Li; Benying Tan; Shuxue Ding; Shengli Xie \| 收藏 \| 浏览/下载：62/0 \| 提交时间：2022/09/08 Difference of convex algorithm (DCA) group log-regularizer key frame extraction structured sparse coding
	Camera-based Basketball Scoring Detection Using Convolutional Neural Network 期刊论文 OAI收割 International Journal of Automation and Computing, 2021, 卷号: 18, 期号: 2, 页码: 266-276 作者: Xu-Bo Fu \| 收藏 \| 浏览/下载：22/0 \| 提交时间：2021/04/22 Computer vision convolutional neural network (CCN) frame difference motion detection object detection real-time system
	Frame difference energy image for gait recognition with incomplete silhouettes 期刊论文 OAI收割 PATTERN RECOGNITION LETTERS, 2009, 卷号: 30, 期号: 11, 页码: 977-984 作者: Chen, Changhong; Liang, Jimin; Zhao, Heng; Hu, Haihong; Tian, Jie 收藏 \| 浏览/下载：56/0 \| 提交时间：2015/08/12 Gait recognition Incomplete silhouettes Frame difference energy image Hidden Markov model
	Research on the affect of differential-images technique to the resolution of infrared spatial camera (EI CONFERENCE) 会议论文 OAI收割 3rd International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test and Measurement Technology and Equipment, July 8, 2007 - July 12, 2007, Chengdu, China Jin G.; An Y.; Qi Y.; Hu F. 收藏 \| 浏览/下载：20/0 \| 提交时间：2013/03/25 The optical system of infrared spatial camera adopts bigger relative aperture and bigger pixel size on focal plane element. These make the system have bulky volume and low resolution. The potential of the optical systems can not be exerted adequately. So one method for improving resolution of infrared spatial camera based on multi-frame difference-images is introduced in the dissertation. The method uses more than one detectors to acquire several difference images and then reconstructs a new high-resolution image from these images through the relationship of pixel grey value. The technique of difference-images that uses more than two detectors is researched and it can improve the resolution 2.5 times in theory. The relationship of pixel grey value between low-resolution difference-images and high-resolution image is found by analyzing the energy of CCD sampling a general relationship between the enhanced times of the resolution of the detected figure with differential method and the least count of CCD that will be used to detect figure is given. Based on the research of theory the implementation process of utilizing difference-images technique to improve the resolution of the figure was simulated used Matlab software by taking a personality image as the object and the software can output the result as an image. The result gotten from the works we have finished proves that the technique is available in high-resolution image reconstruction. The resolution of infrared spatial camera can be improved evidently when holding the size of optical structure or using big size detector by applying for difference image technique. So the technique has a high value in optical remote fields.
	Displacement estimation by the phase-shiftings of fourier transform in present white noise (EI CONFERENCE) 会议论文 OAI收割 ICO20: Optical Information Processing, August 21, 2005 - August 26, 2005, Changchun, China 作者: Wu Y.-H. 收藏 \| 浏览/下载：31/0 \| 提交时间：2013/03/25 Displacement estimation is a fundamental problem in Real-time video image processing. It can be typically approached by theories based on features in spatial domain. This paper presents an algorithm which improves the theory for estimating the moving object's displacement in spatial domain by its Fourier transform frequency spectrum. Because of the characters of Fourier transform the result is based on all the features in the image. Utilizing shift theorem of Fourier transform and auto-registration the algorithm employs the phase spectrum difference in polar coordinate of two frame images sequence with the moving target1 2. The method needn't transform frequency spectrum to spatial domain after calculation comparing with the traditional algorithm which has to search Direc peak and it reduces processing time. Since the technique proposed uses all the image information including all the white noise in the image especially and it's hard to overcome the aliasing from noises but the technique can be an effective way to analyze the result in little white noise by the different characters between high and low frequency bands. It can give the displacement of moving target within 1 pixel of accuracy. Experimental evidence of this performance is presented and the mathematical reasons behind these characteristics are explained in depth. It is proved that the algorithm is fast and simple and can be used in image tracking and video image processing.
	Intelligent MRTD testing for thermal imaging system using ANN (EI CONFERENCE) 会议论文 OAI收割 ICO20: Remote Sensing and Infrared Devices and Systems, August 21, 2005 - August 26, 2005, Changchun, China Sun J.; Ma D. 收藏 \| 浏览/下载：28/0 \| 提交时间：2013/03/25 The Minimum Resolvable Temperature Difference (MRTD) is the most widely accepted figure for describing the performance of a thermal imaging system. Many models have been proposed to predict it. The MRTD testing is a psychophysical task for which biases are unavoidable. It requires laboratory conditions such as normal air condition and a constant temperature. It also needs expensive measuring equipments and takes a considerable period of time. Especially when measuring imagers of the same type the test is time consuming. So an automated and intelligent measurement method should be discussed. This paper adopts the concept of automated MRTD testing using boundary contour system and fuzzy ARTMAP but uses different methods. It describes an Automated MRTD Testing procedure basing on Back-Propagation Network. Firstly we use frame grabber to capture the 4-bar target image data. Then according to image gray scale we segment the image to get 4-bar place and extract feature vector representing the image characteristic and human detection ability. These feature sets along with known target visibility are used to train the ANN (Artificial Neural Networks). Actually it is a nonlinear classification (of input dimensions) of the image series using ANN. Our task is to justify if image is resolvable or uncertainty. Then the trained ANN will emulate observer performance in determining MRTD. This method can reduce the uncertainties between observers and long time dependent factors by standardization. This paper will introduce the feature extraction algorithm demonstrate the feasibility of the whole process and give the accuracy of MRTD measurement.