Part-based adaptive detection of workpieces using differential evolution

Authors:
Wei Liu;Peng Wang;Hong Qiao
Affiliations:
Institute of Automation, Chinese Academy of Sciences, 95 Zhongguancun East Road, 100190 Beijing, China;Institute of Automation, Chinese Academy of Sciences, 95 Zhongguancun East Road, 100190 Beijing, China;Institute of Automation, Chinese Academy of Sciences, 95 Zhongguancun East Road, 100190 Beijing, China
Venue:
Signal Processing
Year:
2012

Citing 8
Cited 1

Hierarchical Chamfer Matching: A Parametric Edge Matching Algorithm

IEEE Transactions on Pattern Analysis and Machine Intelligence
Example-Based Object Detection in Images by Components

IEEE Transactions on Pattern Analysis and Machine Intelligence
Differential Evolution – A Simple and Efficient Heuristic for Global Optimization over Continuous Spaces

Journal of Global Optimization
Shape Matching and Object Recognition Using Shape Contexts

IEEE Transactions on Pattern Analysis and Machine Intelligence
Pictorial Structures for Object Recognition

International Journal of Computer Vision
A Bayesian, Exemplar-Based Approach to Hierarchical Shape Matching

IEEE Transactions on Pattern Analysis and Machine Intelligence
An MRF model-based approach to the detection of rectangular shape objects in color images

Signal Processing
Learned local Gabor patterns for face representation and recognition

Signal Processing

Adaptive object detection by implicit sub-class sharing features

Signal Processing

Quantified Score

Hi-index	0.08

Visualization

Abstract

In many industrial applications, detection of workpieces is the prerequisite of the subsequent operations such as automatic grasping and assembly tasks. However, the detection of workpieces under challenging conditions such as occlusion and cluttered background is still an open problem, which needs better solutions and further investigations. In this paper, a part-based adaptive detection approach is proposed to deal with abovementioned problems. The whole workpiece template is automatically divided into multiple subtemplates, which are equipped with adjustable weights adjusted according to their discriminative abilities. Then the weight adjustment process and the object localization process are finally embedded in an optimization framework-Differential Evolution (DE), which finally leads to the detection of workpieces. Experimental results demonstrate the effectiveness and robust performance of the proposed algorithm under challenging conditions.