Unlock instant, AI-driven research and patent intelligence for your innovation.

Single picture geographic positioning method based on generalized maximum weight map

A geolocation, single image technology, applied in digital data processing, still image data retrieval, special data processing applications, etc., can solve the problems of slow image geolocation matching, low algorithm complexity, low positioning accuracy, etc. Achieve the effect of high positioning accuracy, low algorithm complexity, and improved positioning accuracy

Active Publication Date: 2019-08-16
SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problems of slow matching speed of picture geolocation, local optimal solution and low positioning accuracy, the present invention discloses a single picture geolocation method based on generalized maximum weight map, which has fast matching speed and low positioning accuracy. The result is the global optimal solution and the characteristics of low algorithm complexity

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
  • Single picture geographic positioning method based on generalized maximum weight map
  • Single picture geographic positioning method based on generalized maximum weight map
  • Single picture geographic positioning method based on generalized maximum weight map

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0023] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0024] Such as figure 1 As shown, the single image geolocation method based on the generalized maximum weight map includes a preprocessing part and a real-time matching part. For the reference picture set, this specific embodiment adopts the 102K street view data set and the 300K world-class city data set disclosed by Zamir and Zemene, which respectively correspond to 644 and 500 test pictures with GPS tags. Such as figure 2 Shown is a visualization of the main steps required in the process of locating a query image.

[0025] Wherein the preprocessing part is to extract the local feature descriptor of the street view image as a reference feature, establish a mapping table between the local feature point and the extracted street view image, and organize the reference feature set into an index structure; the specific steps are as follows:

[...

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 discloses a single picture geographic positioning method based on a generalized maximum weight map. The method comprises the following steps: firstly, extracting a local descriptor froma query picture as a query feature; secondly, retrieving nearest neighbors for each query feature in the index of the pre-processed organization and roughly removing non-significant query features andcorresponding nearest neighbors by using a dynamic pruning method; next step, formalizing the geographic positioning problem into a matching problem between the query picture and the candidate picture; namely, regarding the reference picture corresponding to the nearest neighbor as a potential matching picture, trimming candidate pictures with low matching degree and selecting a picture with thehighest matching similarity from the candidate pictures by using a picture matching method based on a generalized maximum weight picture to serve as a final result. The method has the characteristicsof high positioning precision, low algorithm complexity and the like.

Description

technical field [0001] The invention relates to the technical field of image positioning, in particular to a method for geographic positioning of a single picture based on a generalized maximum weight map. Background technique [0002] The geolocation of pictures plays an important role in images and videos, and they have many applications in the real world. In the field of auxiliary analysis, geolocation can be used to organize image collections, statistical analysis of law enforcement and commercial market trends, etc.; in specific technical fields, it can be used for model-based image enhancement, augmented reality, vehicle video trajectory estimation and self-driving cars, etc. ; In the field of intelligent transportation, it can be used for combined positioning technology based on vision-based geographic positioning and other non-visual positioning systems, which can achieve higher robustness, real-time and accuracy. [0003] In the geolocation method of large-scale st...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/58G06F16/587
CPCG06F16/58G06F16/587
Inventor 张凯邓文泉董宇涵
Owner SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV