Descriptor generation method and apparatus

A descriptor and main direction technology, applied in the field of descriptor generation methods and devices, can solve the problems of poor descriptor accuracy, large difference in local area texture structure, low multi-mode image registration accuracy, etc. The effect of improving registration accuracy

Active Publication Date: 2015-10-14
BEIJING UNIV OF POSTS & TELECOMM
View PDF2 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, if the image registration is a multimodal image registration, since the texture structure of the local area of ​​the registered image is quite different, the main directions of the corresponding feature points of each image obtained according to the gradient information are quite different, and the description The

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Descriptor generation method and apparatus
  • Descriptor generation method and apparatus
  • Descriptor generation method and apparatus

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0041] figure 1 It is a flowchart of Embodiment 1 of the descriptor generation method of the present invention. The execution subject of this embodiment is a descriptor generating device, which is suitable for scenarios where descriptors need to be generated in multi-mode image registration. Specifically, this embodiment includes the following steps:

[0042] 101. Extract feature points in the image.

[0043] In this step, the feature point detector is used to detect the feature points in the image. The feature point detector can be, for example, Fast Feature Detector (FAST), Fast Retina Key (FREAK) detector, SIFT detection 器等。

[0044] 102. Extract the contour of the image.

[0045] The image in the embodiment of the present invention refers to the image to be registered. Steps 101 to 105 are performed on each image to generate a descriptor. Then, in the multi-mode image registration, the descriptor is used to compare each image according to a certain similarity criterion. The corr...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a descriptor generation method and apparatus. The method comprises the steps of: extracting the profile of an input image, and extracting at least one line segment according to the profile; determining a main direction of a detected characteristic point according to the direction of the line segment; rotating a neighborhood of the characteristic point according to the main direction; and describing the characteristic point according to the rotated neighborhood so as to generate a descriptor of the characteristic point. In the process, the determination of the main direction is mainly dependent on the direction of the line segment of the neighborhood of the characteristic point, and in a multimodal image, the line segment is more stable relative to the point, so that the descriptor of the characteristic point corresponding to each image has repeatability, the accuracy of the descriptor is improved, and further the registration accuracy of the multimodal image is improved.

Description

technical field [0001] The invention relates to image registration technology, in particular to a descriptor generation method and device. Background technique [0002] Image registration refers to the process of matching and superimposing at least two images acquired at different times, different imaging sensors or under different conditions. In this process, if the images to be registered come from the same type of imaging sensors, such as visible light imaging sensors or infrared imaging sensors, the registration process is called single-mode image registration; otherwise, if the registered The images come from different types of imaging sensors, and the registration process is called multimodal image registration. [0003] In existing image registration based on feature points, feature points are extracted for each image, and the main direction of the feature points is determined, and the coordinate axis is rotated to the main direction of the feature points to ensure r...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06T3/00G06T3/60
CPCG06T3/0068G06T3/60
Inventor 李永温志刚刘杰余杭范若琛魏丽君徐良鹏
Owner BEIJING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products