中国科学院机构知识库网格系统: 检索

Location Sensitive Network for Human Instance Segmentation 期刊论文 OAI收割
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2021, 卷号: 30, 页码: 7649-7662
作者: Zhang, Xiangzhou; Ma, Bingpeng; Chang, Hong; Shan, Shiguang; Chen, Xilin

|

收藏 | 浏览/下载：25/0 | 提交时间：2021/12/01

Image segmentation Prototypes Heating systems Task analysis Semantics Feature extraction Detectors Human instance segmentation spatial invariance coordinates encoding points representation

KCCA-based radiation normalization method for hyperspectral remote sensing Images 会议论文 OAI收割
Beijing, China, 2019-07-07
作者: Li, Haiwei; Song, Liyao; Yan, Qiangqiang; Chen, Tieqiao

|

收藏 | 浏览/下载：55/0 | 提交时间：2020/03/04

Kernel Canonical Correlation Analysis (KCCA) Radiation normalization Pseudo-invariant feature points (PIF) Multivariate change detection

Hybrid patch based diagonal pattern geometric appearance model for facial expression recognition 会议论文 OAI收割
Beijing, 2016-10-19
作者: Jain Deepak Kumar; Zhang Zhang; Huang Kaiqi

|

收藏 | 浏览/下载：29/0 | 提交时间：2018/06/19

Facial Expressions Feature Points

Measurement algorithm of mounting holes based on binocular stereo vision 会议论文 OAI收割
6th Annual IEEE International Conference on Cyber Technology in Automation, Control and Intelligent Systems, IEEE-CYBER 2016, Chengdu, China, June 19-22, 2016
作者: Hu YQ(胡彦强); Ma Y(马钺); Xu M(许敏); Chen X(陈曦)

收藏 | 浏览/下载：28/0 | 提交时间：2016/11/06

epipolar-line constraint algorithm feature points matching Line laser mounting holes

A simple pose estimation method based on co-planar feature points 会议论文 OAI收割
control, automation and robotics (iccar), 2015 international conference on, singapore, 2015-05
作者: Weigao Xu; Zuofeng Zhou; Jianzhong Cao; Bing Zhao

收藏 | 浏览/下载：19/0 | 提交时间：2016/01/08

P4P closed-form solution co-planar feature points pose estimation

Data normalization of single camera visual measurement network system (EI CONFERENCE) 会议论文 OAI收割
2012 International Conference on Information Technology and Management Innovation, ICITMI 2012, November 10, 2012 - November 11, 2012, Guangzhou, China
作者: Zhang Y.-C.; Zhou J.

收藏 | 浏览/下载：23/0 | 提交时间：2013/03/25

With the development of visual measurement system Normalize the measured coordinates into the same world coordinate system Apply in the three coordinate measuring machine(CMM) to have the simulation experiment So the data normalization has the advantage of more high precision. (2013) Trans Tech Publications the visual measurement system of single camera which applied in the imaging theory of optical feature points then get the global data the result indicates that the maximum absolute tolerance between the normalization coordinates via the center of point set and the ones measured by CMM directly is 0.058mm Switzerland. it has been widespread used in the modern production. Due to the limit of the environment in scene and achieve the overall measurement the visual measurement system of single camera could not measure the shield between the measured objects each other. Focus on this problem But the one between the coordinate repeatedly measured at the position of one network control point is 0.066mm present a kind of the the knowledge of measurement network based on the visual measurement of single camera set up the measurement network system via the multi-control points. Measure the optical feature points in every network control point via the visual measurement system of single camera

Image matching combine SIFT with regional SSDA (EI CONFERENCE) 会议论文 OAI收割
2012 International Conference on Control Engineering and Communication Technology, ICCECT 2012, December 7, 2012 - December 9, 2012, Shenyang, Liaoning, China
作者: Liu J.; Liu J.; Liu J.; Zhao J.

收藏 | 浏览/下载：37/0 | 提交时间：2013/03/25

Image matching is at the base of many computer vision problems such as object recognition or image stitching. Standard SIFT provides poor performance when images under viewpoint change conditions and with similar corners. Hence we propose a matching algorithm combine regional SSDA with simplified SIFT algorithm. We demonstrate through experiments that our algorithm yields better performance in images of viewpoint change and similar feature points. Besides the simplified algorithm cut down about half the time was originally needed in our tested images. 2012 IEEE.

Image registration based on Mexican-hat wavelets and pseudo-Zernike moments (EI CONFERENCE) 会议论文 OAI收割
2012 World Automation Congress, WAC 2012, June 24, 2012 - June 28, 2012, Puerto Vallarta, Mexico
作者: Liu Y.; Liu Y.; Liu Y.

收藏 | 浏览/下载：36/0 | 提交时间：2013/03/25

Image registration is a key technique in pattern recognition and image processing and it is widely used in many application areas such as computer vision remote sensing image fusion and object tracking. A method for image registration combining Mexican-hat wavelets and pseudo-Zernike moments is proposed. Firstly feature points are extracted using scale-interaction Mexican-hat wavelets in the reference image and sensed image respectively. Then pseudo-Zernike moments are used to match them and classical RANSAC used to eliminate the wrong matches. And then the well match points are used to estimate the best affine transform parameters by least squares minimization. At last the sensed image is transformed and resampled to accomplish the image registration. The experiments indicate that the proposed algorithm extracts feature points and matches them exactly and eliminates wrong matched points effectively and achieves nice registration results. 2012 TSI Press.

Features extraction and matching of teeth image based on the SIFT algorithm (EI CONFERENCE) 会议论文 OAI收割
2012 2nd International Conference on Computer Application and System Modeling, ICCASM 2012, July 27, 2012 - July 29, 2012, Shenyang, China
作者: Wang X.; Wang X.; Wang X.

收藏 | 浏览/下载：28/0 | 提交时间：2013/03/25

Using of SIFT algorithm in the image of teeth model can detect the features of the teeth image effectively. In this approach first search over all scales and image locations by using a difference-of-Gaussian function to identify potential interest points that are invariant to scale and orientation. Second select keypoints based on measures of their stability and a detailed model is fit to determine location and scale at each candidate location. Third assign one or more orientations to each keypoint location based on local image gradient directions. Last measure the local image gradients at the selected scale in the region around each keypoint. And then use the KNN algorithm to match the features. Through lots of experiments and comparing with other feature extraction methods this method can detect the features of the teeth model effectively and offer some available parameters for 3D reconstruction of the teeth model. the authors.

The aircraft autopilot design of parameters changing in a wide range (EI CONFERENCE) 会议论文 OAI收割
2011 International Conference on Electric Information and Control Engineering, ICEICE 2011, April 15, 2011 - April 17, 2011, Wuhan, China
Zhen D.; Yue Z.

收藏 | 浏览/下载：28/0 | 提交时间：2013/03/25

Autopilot is an important part of aircraft Under the action of the disturbance torque for the flight parameters of aircraft Changing in a wide range the range was about 16km on the basic of analyzing the typical trajectory the max height was about 5100m the trajectory of feature points were selected the deviation of lateral was about 12m taking pitch control loop for example the maximum angle of attack was about 3.6 a sub- PID autopilot was designed using classical control theory the maximum elevator deflection was less than 5 using single-mode Transient suppression method the maximum aileron deflection was less than 2 the overload due to control surface transition deflection was avoided when control parameters changing the maximum roll was less than 5 the verification was done by non-linear simulation. The result showed that: the phase margin of autopilot was greater than 30 the response of overload can accurately track the given overload instruction. An autopilot design method of aircraft of parameter changes in a wide range was provided. 2011 IEEE. the bandwidth was more than 14rad / s the settling time was less than 0.7s the maximum overshoot was 23.3%