Object tracking based on local feature points

2010 ◽  
Author(s):  
Haili Wang ◽  
Liang Zhang
Keyword(s):  
Object Tracking ◽  
Local Feature ◽  
Feature Points

Seam tracking investigation via striped line laser sensor

2017 ◽  
Vol 44 (5) ◽  
pp. 609-617 ◽  
Author(s):  
Yanbiao Zou ◽  
Jinchao Li ◽  
Xiangzhi Chen
Keyword(s):  
Machine Vision ◽  
Object Tracking ◽  
Seam Tracking ◽  
Cosine Similarity ◽  
Feature Points ◽  
Robot Arm ◽  
Coordinate Systems ◽  
Welding Seam ◽  
Content Type ◽  
Experiment Platform

Purpose This paper aims to propose a set of six-axis robot arm welding seam tracking experiment platform based on Halcon machine vision library to resolve the curve seam tracking issue. Design/methodology/approach Robot-based and image coordinate systems are converted based on the mathematical model of the three-dimensional measurement of structured light vision and conversion relations between robot-based and camera coordinate systems. An object tracking algorithm via weighted local cosine similarity is adopted to detect the seam feature points to prevent effectively the interference from arc and spatter. This algorithm models the target state variable and corresponding observation vector within the Bayes framework and finds the optimal region with highest similarity to the image-selected modules using cosine similarity. Findings The paper tests the approach and the experimental results show that using metal inert-gas (MIG) welding with maximum welding current of 200A can achieve real-time accurate curve seam tracking under strong arc light and splash. Minimal distance between laser stripe and welding molten pool can reach 15 mm, and sensor sampling frequency can reach 50 Hz. Originality/value Designing a set of six-axis robot arm welding seam tracking experiment platform with a system of structured light sensor based on Halcon machine vision library; and adding an object tracking algorithm to seam tracking system to detect image feature points. By this technology, this system can track the curve seam while welding.

scholarly journals Building Detection from VHR Remote Sensing Imagery Based on the Morphological Building Index

2018 ◽  
Vol 10 (8) ◽  
pp. 1287 ◽  
Author(s):  
Yongfa You ◽  
Siyuan Wang ◽  
Yuanxu Ma ◽  
Guangsheng Chen ◽  
Bin Wang ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Spectral Characteristics ◽  
Local Feature ◽  
Building Detection ◽  
Feature Points ◽  
Complex Environments ◽  
Remote Sensing Imagery ◽  
Complex Image ◽  
Very High ◽  
A Current

Automatic detection of buildings from very high resolution (VHR) satellite images is a current research hotspot in remote sensing and computer vision. However, many irrelevant objects with similar spectral characteristics to buildings will cause a large amount of interference to the detection of buildings, thus making the accurate detection of buildings still a challenging task, especially for images captured in complex environments. Therefore, it is crucial to develop a method that can effectively eliminate these interferences and accurately detect buildings from complex image scenes. To this end, a new building detection method based on the morphological building index (MBI) is proposed in this study. First, the local feature points are detected from the VHR remote sensing imagery and they are optimized by the saliency index proposed in this study. Second, a voting matrix is calculated based on these optimized local feature points to extract built-up areas. Finally, buildings are detected from the extracted built-up areas using the MBI algorithm. Experiments confirm that our proposed method can effectively and accurately detect buildings in VHR remote sensing images captured in complex environments.

Smooth Edge Feature Lines Extraction from Point Clouds of Eroded Fractured Fragments

2013 ◽  
Vol 756-759 ◽  
pp. 4026-4030 ◽  
Author(s):  
Jian Bin Lin ◽  
Ming Quan Zhou ◽  
Zhong Ke Wu
Keyword(s):  
Principal Component Analysis ◽  
Principal Component ◽  
Point Clouds ◽  
Local Feature ◽  
Feature Points ◽  
Global Feature ◽  
Smooth Edge ◽  
Feature Lines ◽  
Novel Method ◽  
Edge Feature

This paper presents a novel method to extract edge lines from point clouds of these eroded, rough fractured fragments. Firstly, a principal component analysis based method is used to extract feature points, followed by clustering of these feature points. Secondly, a local feature lines fragment is constructed for each cluster and afterwards a smooth and noise pruning process for each local feature lines fragment. Thirdly, these separated local feature lines fragments are connected and bridged in order to eliminate the gaps caused by the eroded regions and construct completed global feature lines. Fourthly, a noise pruning process is performed. The output of this method is completed, smoothed edge feature lines. We illustrate the performance of our method on a number of real-world examples.

scholarly journals Fundus Image Registration Technique Based on Local Feature of Retinal Vessels

2021 ◽  
Vol 11 (23) ◽  
pp. 11201
Author(s):  
Roziana Ramli ◽  
Khairunnisa Hasikin ◽  
Mohd Yamani Idna Idris ◽  
Noor Khairiah A. Karim ◽  
Ainuddin Wahid Abdul Wahab
Keyword(s):  
Feature Extraction ◽  
Image Registration ◽  
Extraction Methods ◽  
Local Feature ◽  
Retinal Vessels ◽  
Feature Points ◽  
Fundus Image ◽  
Scale Invariant ◽  
Feature Based ◽  
Image Pairs

Feature-based retinal fundus image registration (RIR) technique aligns fundus images according to geometrical transformations estimated between feature point correspondences. To ensure accurate registration, the feature points extracted must be from the retinal vessels and throughout the image. However, noises in the fundus image may resemble retinal vessels in local patches. Therefore, this paper introduces a feature extraction method based on a local feature of retinal vessels (CURVE) that incorporates retinal vessels and noises characteristics to accurately extract feature points on retinal vessels and throughout the fundus image. The CURVE performance is tested on CHASE, DRIVE, HRF and STARE datasets and compared with six feature extraction methods used in the existing feature-based RIR techniques. From the experiment, the feature extraction accuracy of CURVE (86.021%) significantly outperformed the existing feature extraction methods (p ≤ 0.001*). Then, CURVE is paired with a scale-invariant feature transform (SIFT) descriptor to test its registration capability on the fundus image registration (FIRE) dataset. Overall, CURVE-SIFT successfully registered 44.030% of the image pairs while the existing feature-based RIR techniques (GDB-ICP, Harris-PIIFD, Ghassabi’s-SIFT, H-M 16, H-M 17 and D-Saddle-HOG) only registered less than 27.612% of the image pairs. The one-way ANOVA analysis showed that CURVE-SIFT significantly outperformed GDB-ICP (p = 0.007*), Harris-PIIFD, Ghassabi’s-SIFT, H-M 16, H-M 17 and D-Saddle-HOG (p ≤ 0.001*).

Sign in / Sign up

Export Citation Format

Share Document