image registration plays an important role in many practical problems in diverse fields such as medical, remote sensing, andcomputervision. The basic purpose of image registration is to obtain the finest geometrical...
详细信息
ISBN:
(纸本)9781509059546
image registration plays an important role in many practical problems in diverse fields such as medical, remote sensing, andcomputervision. The basic purpose of image registration is to obtain the finest geometrical and radio-metrically aligned image from temporal or multi-modal image sensors. In this paper, a novel stable shape feature-based image registration method has been proposed by matching the stable region with a set of rotations, scale invariant features, and multi-scale image segmentation is used to obtain the matching areas. This algorithm firstly converts images into image objects by multi-scale segmentation and convexity model restriction. Then these reliable and stable image regions are used as matching unit rather than points or lines. Experiments show the algorithm proposed in this paper is not sensitive to rotation and resolution distortion, which can accomplish image registration automatically.
In this paper, a work in progress towards a real-time vision-based traffic flow prediction (TFP) system is presented. The proposed method consists of three elemental operators, that are dynamic texture model based mot...
详细信息
ISBN:
(纸本)9781509059546
In this paper, a work in progress towards a real-time vision-based traffic flow prediction (TFP) system is presented. The proposed method consists of three elemental operators, that are dynamic texture model based motion segmentation, feature extraction and Gaussian process (GP) regression. The objective of motion segmentation is to recognize the target regions covering the moving vehicles in the sequence of visual processes. The feature extraction operator aims to extract useful features from the target regions. The extracted features are then mapped to the number of vehicles through the operator of GP regression. A training stage using historical visual data is required for determining the parameter values of the GP. Based on a low-resolution visual data set, preliminary evaluations on the performance of the proposed method are performed. The results show that the proposed method beats a benchmark solution based on Gaussian mixture model, and has the potential to be developed into qualified and practical solutions to real-time TFP.
This paper focuses on a task of automatic pedestrian detection. There arc many possible solutions with several ideas behind. Histogram of oriented gradients is among the popular ones. The former is used as a descripto...
详细信息
ISBN:
(纸本)9781538651636
This paper focuses on a task of automatic pedestrian detection. There arc many possible solutions with several ideas behind. Histogram of oriented gradients is among the popular ones. The former is used as a descriptor of the input image together with support vector machine. In this initial study, we propose to create a feature vector using fuzzy mathematics and test its effectiveness. The effectiveness is evaluated by comparison with original histogram of oriented gradient descriptors in support vector machine.
image captioning is a crucial task in the interaction of computervision and natural language processing. It is an important way that help human understand the world better. There are many studies on image English cap...
详细信息
ISBN:
(纸本)9781509044993
image captioning is a crucial task in the interaction of computervision and natural language processing. It is an important way that help human understand the world better. There are many studies on image English captioning, but little work on image Chinese captioning because of the lack of the corresponding datasets. This paper focuses on image Chinese captioning by using abundant English datasets for the issue. In this paper, a method of adding English information to image Chinese captioning is proposed. We validate the use of English information with state-of-the art performance on the datasets: Flickr8K-CN.
vision sensors are attractive for autonomous robots because they give huge source of environment information. Obstacle detection is main task in areas of the robotics, vision based works gives the effective performanc...
详细信息
ISBN:
(纸本)9781509043071
vision sensors are attractive for autonomous robots because they give huge source of environment information. Obstacle detection is main task in areas of the robotics, vision based works gives the effective performance imageprocessing such as edge detection techniques are used to make images more effective and methods to implement detection process for increasing the performance of detecting obstacles in indoor environment. Omni directional vision gives panoramic view that has benefit of 360 degree so as to detect obstacles behind robot. image Segmentation is used to segment the environment into regions done by GMM on each region the method is applied for detecting the obstacles. imageprocessing and classification methods are used to avoid obstacles and navigate the Safe path for autonomous robots.
The presence of atmospheric turbulence over horizontal imaging paths introduces time-varying perturbations and blur in the scene that severely degrade the performance of moving object detection and tracking systems of...
详细信息
Embedded image acquisition, processing and transmission systems have advantages of small volume and high stability, etc, and therefore they are widely used in intelligent transportation, electric power, communication ...
详细信息
ISBN:
(纸本)9781424445189
Embedded image acquisition, processing and transmission systems have advantages of small volume and high stability, etc, and therefore they are widely used in intelligent transportation, electric power, communication andcomputervision, etc. Along with the development of DSP technology, realizing real-time video compression on DSP by software has become a highlight of digital video compression standard application and this kind of application has greater flexibility and upgrading potential than special compression chips. The continuous development of modern industrial technology has put forward higher requirements for the acquisition, processing and transmission of image information. In recent years, high-speed digital signal processing DSP technology develops rapidly and the strong signal processing ability of DSP has provided solid theoretical and application foundation for the real-time processing of image information. First, this paper discusses the hardware structure of image tracking system taking ARM-DSP as main frame and the development process and control flow of DSP. Finally, it looks forward to the development prospect of imageprocessing. This paper studies basic theories of digital imageprocessing.
Storing paper-based documents converted into digital image format is one of effective solutions to preserve content in a document. Searching the Document images stored in a repository by using time as queries is one o...
详细信息
ISBN:
(纸本)9781538609699
Storing paper-based documents converted into digital image format is one of effective solutions to preserve content in a document. Searching the Document images stored in a repository by using time as queries is one of importance tasks of document retrieval. This paper presents a simple scheme of imageprocessing to retrieve the documents, which contain the desired date of issue printed in Thai alphabets, from a repository. The procedure of this retrieval scheme consists of 4 stages: image acquisition, pre-processing, zone identification, and pattern recognition. The proposed machine vision scheme gave the excellent result over 365 dummy dates of issue of documents in the tested repository. It is a simple and uncomplex approach for the document image retrieval of Thai government letters, printed in Thai alphabets, using date of issue as query.
In recent years, machine vision technology has been widely used in production. The scanners of the three-dimensional scanning technology are too expensive to promotion. A new height measurement method of irregular obj...
详细信息
ISBN:
(纸本)9781509059546
In recent years, machine vision technology has been widely used in production. The scanners of the three-dimensional scanning technology are too expensive to promotion. A new height measurement method of irregular objects is proposed in this paper. Firstly, the color and depth images of objects are collected by using Kinect, then the images collected are needed to preprocessed. The height matrix is obtained by background subtraction method. Finally, the height matrix element values are divided into different interval and then the height of the object is calculated by finding the maximum connected domain. The experimental results show that the relative error is less than 1.56%, which improves the efficiency and accuracy of height measurement, and it realizes the automation of the measurement process and can satisfy the requirement of real-time.
Histogram-based image enhancement methods are widely used in image pre-processing, which are vital for the consequential image analysis, feature extraction and object recognition. This paper studies six adaptive metho...
详细信息
ISBN:
(纸本)9781509062386
Histogram-based image enhancement methods are widely used in image pre-processing, which are vital for the consequential image analysis, feature extraction and object recognition. This paper studies six adaptive methods which modify the conventional histogram equalization (HE) and further concern the local quality of the normalized images. The parameters for controlling the normalization effects are also defined and discussed. During this study, we also show the relation between these methods. Experiments are conducted to evaluate the performance of these methods and the results using various standard quantative measures are also given.
暂无评论