A Fast Template Matching Algorithm Based on Wavelet Coefficient Projection Transform

2011 ◽  
Vol 48-49 ◽  
pp. 21-24 ◽  
Author(s):  
Xian Ping Fu ◽  
Sheng Long Liao
Keyword(s):  
Computer Vision ◽  
Data Structure ◽  
Template Matching ◽  
Performance Enhancement ◽  
Wavelet Coefficient ◽  
Target Position ◽  
Similarity Matching ◽  
Structure Algorithm ◽  
Searching Ability ◽  
Resolution Data

As the electronic industry advances rapidly toward automatic manufacturing smaller, faster, and cheaper products, computer vision play more important role in IC packaging technology than before. One of the important tasks of computer vision is finding target position through similarity matching. Similarity matching requires distance computation of feature vectors for each target image. In this paper we propose a projection transform of wavelet coefficient based multi resolution data-structure algorithm for faster template matching, a position sequence of local sharp variation points in such signals is recorded as features. The proposed approach reduces the number of computation by around 70% over multi resolution data structure algorithm. We use the proposed approach to match similarity between wavelet parameters histograms for image matching. It is noticeable that the proposed fast algorithm provides not only the same retrieval results as the exhaustive search, but also a faster searching ability than existing fast algorithms. The proposed approach can be easily combined with existing algorithms for further performance enhancement.

WAVELET-BASED MULTIRESOLUTION HISTOGRAM FOR FAST IMAGE RETRIEVAL

2004 ◽  
Vol 02 (01) ◽  
pp. 59-73 ◽  
Author(s):  
PAWAN JAIN ◽  
S. N. MERCHANT
Keyword(s):  
Data Structure ◽  
Image Retrieval ◽  
Retrieval System ◽  
Performance Enhancement ◽  
Image Database ◽  
Retrieval Systems ◽  
Candidate Image ◽  
Image Retrieval System ◽  
Resolution Level ◽  
Structure Algorithm

Most of the content-based image retrieval systems require a distance computation of feature vectors for each candidate image in the image database. This exhaustive search is highly time-consuming and inefficient. This limits the usefulness of such system. Thus there is a growing need for a fast image retrieval system. Multiresolution data-structure algorithm provides a good solution to the above problem. In this paper we propose a wavelet-based multiresolution data-structure algorithm. Wavelet-based multiresolution data-structure further reduce the number of computation by around 50%. In the proposed approach we reuse the information obtained at lower resolution levels to calculate the distance at a higher resolution level. Apart from this, the proposed structure saves memory overheads by about 50% over multiresolution data-structure algorithm. The proposed algorithm can be easily combined with other algorithms for performance enhancement.4 In this paper we use the proposed technique to match luminance histogram for image retrieval. Fuzzy histograms enhances performance by considering the similarity between neighboring bins. We have extended the proposed approach to fuzzy histograms for better performance.

scholarly journals Random Forest with Adaptive Local Template for Pedestrian Detection

2015 ◽  
Vol 2015 ◽  
pp. 1-11 ◽  
Author(s):  
Tao Xiang ◽  
Tao Li ◽  
Mao Ye ◽  
Zijian Liu
Keyword(s):  
Computer Vision ◽  
Random Forest ◽  
Classification Accuracy ◽  
Template Matching ◽  
Detection Method ◽  
State Of The Art ◽  
Pedestrian Detection ◽  
Sliding Window ◽  
Experimental Results ◽  
Training Samples

Pedestrian detection with large intraclass variations is still a challenging task in computer vision. In this paper, we propose a novel pedestrian detection method based on Random Forest. Firstly, we generate a few local templates with different sizes and different locations in positive exemplars. Then, the Random Forest is built whose splitting functions are optimized by maximizing class purity of matching the local templates to the training samples, respectively. To improve the classification accuracy, we adopt a boosting-like algorithm to update the weights of the training samples in a layer-wise fashion. During detection, the trained Random Forest will vote the category when a sliding window is input. Our contributions are the splitting functions based on local template matching with adaptive size and location and iteratively weight updating method. We evaluate the proposed method on 2 well-known challenging datasets: TUD pedestrians and INRIA pedestrians. The experimental results demonstrate that our method achieves state-of-the-art or competitive performance.

A Data Abstraction Alternative to Data Structure/Algorithm Modularization

2000 ◽  
pp. 102-113 ◽  
Author(s):  
Murali Sitaraman ◽  
Bruce W. Weide ◽  
Timothy J. Long ◽  
William F. Ogden
Keyword(s):  
Data Structure ◽  
Data Abstraction ◽  
Structure Algorithm

scholarly journals Fashion Product Classification through Deep Learning and Computer Vision

2019 ◽  
Vol 9 (7) ◽  
pp. 1385 ◽  
Author(s):  
Luca Donati ◽  
Eleonora Iotti ◽  
Giulio Mordonini ◽  
Andrea Prati
Keyword(s):  
Image Processing ◽  
Computer Vision ◽  
Feature Extraction ◽  
Deep Learning ◽  
Template Matching ◽  
Learning Approaches ◽  
Visual Classification ◽  
Product Classification ◽  
Processing Techniques

Visual classification of commercial products is a branch of the wider fields of object detection and feature extraction in computer vision, and, in particular, it is an important step in the creative workflow in fashion industries. Automatically classifying garment features makes both designers and data experts aware of their overall production, which is fundamental in order to organize marketing campaigns, avoid duplicates, categorize apparel products for e-commerce purposes, and so on. There are many different techniques for visual classification, ranging from standard image processing to machine learning approaches: this work, made by using and testing the aforementioned approaches in collaboration with Adidas AG™, describes a real-world study aimed at automatically recognizing and classifying logos, stripes, colors, and other features of clothing, solely from final rendering images of their products. Specifically, both deep learning and image processing techniques, such as template matching, were used. The result is a novel system for image recognition and feature extraction that has a high classification accuracy and which is reliable and robust enough to be used by a company like Adidas. This paper shows the main problems and proposed solutions in the development of this system, and the experimental results on the Adidas AG™ dataset.

Recognition of Vehicles on Geometric Morphology

2011 ◽  
Vol 217-218 ◽  
pp. 27-32
Author(s):  
Guo Feng Qin ◽  
Yu Sun ◽  
Qi Yan Li
Keyword(s):  
Computer Vision ◽  
Pattern Recognition ◽  
Feature Extraction ◽  
Traffic Management ◽  
Template Matching ◽  
Recognition System ◽  
Automatic Recognition ◽  
Vehicular Traffic ◽  
Image Pretreatment ◽  
Geometric Morphology

Detection of vehicles plays an important role in the area of the modern intelligent traffic management. And the pattern recognition is a hot issue in the area of computer vision. This article introduces an Automobile Automatic Recognition System based on image. It begins with the structures of the system. Then detailed methods for implementation are discussed. This system take use of a camera to get traffic images, then after image pretreatment and segmentation, do the works of feature extraction, template matching and pattern recognition, to identify different models and get vehicular traffic statistics. Finally, the implementation of the system is introduced. The algorithms of recognized process were verified in this application case.

scholarly journals The GeoHashTree: a multi-resolution data structure for the management of point clouds

2014 ◽  
Author(s):  
N Sabo ◽  
A Beaulieu ◽  
D Bélanger ◽  
Y Belzile ◽  
B Piché
Keyword(s):  
Data Structure ◽  
Point Clouds ◽  
Resolution Data

scholarly journals Development of an AI-based bot to Tibia MMORPG

2021 ◽  
Vol 2 (2) ◽  
pp. 1-9
Author(s):  
Thiago Castanheira Retes de Sousa ◽  
Rafael Lima de Carvalho
Keyword(s):  
Artificial Intelligence ◽  
Computer Vision ◽  
Template Matching ◽  
Search Algorithm ◽  
Real Life ◽  
Input Image ◽  
Graph Search ◽  
As Graph ◽  
Simulation Results ◽  
And Performance

Artificial Intelligence has always been used in designing of automated agents for playing games such as Chess, Go, Defense of the Ancients 2, Snake Game, billiard and many others. In this work, we present the development and performance evaluation of an automated bot that mimics a real life player for the RPG Game Tibia. The automated bot is built using a combination of AI techniques such as graph search algorithm A* and computer vision tools like template matching. Using four algorithms to get global position of player in game, handle its health and mana, target monsters and walk through the game, we managed to develop a fully automated Tibia bot based in raw input image. We evaluated the performance of the agent in three different scenarios, collecting and analyzing metrics such as XP Gain, Supplies Usage and Balance. The simulation results shows that the developed bot is capable of producing competitive results according to in-game metrics when compared to human players.

scholarly journals Use of Dynamic Time Warping for Object Shape Classification through Signature

1970 ◽  
Vol 6 (1) ◽  
pp. 33-49 ◽  
Author(s):  
KC Santosh
Keyword(s):  
Template Matching ◽  
Dynamic Time Warping ◽  
Center Of Gravity ◽  
Shape Classification ◽  
Time Warping ◽  
Similarity Matching ◽  
Object Shape ◽  
Boundary Points ◽  
Engineering And Technology ◽  
Dynamic Time

This paper expresses an application of similarity matching of the signatures through DTW.Fundamental aspect of classification is template matching. The classification is robust tonoise, scaling, and rotation. Feature includes radius plus angle along the boundary points withrespect to center of gravity. The classification automatically and confidently discloses theshape of every object at once throughout page from top to bottom. The paper expresses itspromising results within an average of a few seconds (cheaper classification) for an object. Aseries of tests is done with all possible configurations of geometrical shapes.Keywords: Signature; Dynamic Time Warping; Uniform ScalingDOI: 10.3126/kuset.v6i1.3308Kathmandu University Journal of Science, Engineering and Technology Vol.6(1) 2010, pp33-49

Sign in / Sign up

Export Citation Format

Share Document