Geographical annotation method, device and computer-readable storage medium

A technology based on region and calculation formula, applied in computing, instrument, multimedia data retrieval, etc., can solve the problems of uneven quality of outsourcing personnel, waste of manpower, large errors, etc., to achieve the effect of improving labeling accuracy and labeling efficiency, and reducing information inconsistency

Active Publication Date: 2019-02-15
XIAMEN MEITUZHIJIA TECH
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The common image region labeling mainly relies on manual work, which is handed over to the operators of the image community to mark the regions of popular images by understanding the text information, which has high accuracy but wastes manpower; it is handed over to other image outsourcing personnel to carry out image region labeling Labeling, due to the uneven quality of outsourcing personnel, the standards for regional labeling of pictures are different, the error is large, and resources are wasted, which can also easily lead to data leakage

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
  • Geographical annotation method, device and computer-readable storage medium
  • Geographical annotation method, device and computer-readable storage medium
  • Geographical annotation method, device and computer-readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] This embodiment provides a region labeling method, such as figure 1 As shown, the method includes:

[0069] S101. Acquire low-weight words and high-weight words in a preset data set, wherein a first weight value of a low-weight word is smaller than a first preset threshold, and a second weight value of a high-weight word is larger than a second preset threshold.

[0070] Wherein, in this step, the preset data set includes multiple documents, and each document has its corresponding regional label. The regional label is used to mark the regional information described by the current document information. The regional label can include multiple levels according to the level of the region. Regional labels, for example, when there are two levels of regional labels, the first-level regional labels can be provinces or municipalities directly under the central government, and the second-level regional labels are corresponding famous cities below the provinces. The first-level re...

Embodiment 2

[0138] This embodiment provides a region labeling method, such as Image 6 As shown, this embodiment will be specifically described below in combination with the practical application of the method.

[0139] Step 601. Obtain the low-weight words and high-weight words in the preset data set. That is, assuming that the preset data set includes multiple documents, each document has its corresponding regional label, and the regional label includes two first-level regional labels, namely "Beijing" and "Fujian", where "Fujian" includes two second-level The geographical labels are "Xiamen" and "Fuzhou". Here, the word segmentation operation is performed on the documents in the preset data set according to the preset word segmentation rules, and the high-weight words in the document with the region label "Beijing" are "Beijing" and "Tiananmen"; the region label is "Fujian" The high-weight words in the documents with the region label "Xiamen" are "Xiamen" and "Gulangyu", and the high...

Embodiment 3

[0147] This embodiment provides a region labeling device 300. It should be noted that the basic principles and technical effects of the region labeling device 300 provided in this embodiment are the same as those of the aforementioned corresponding method embodiments. For a brief description, this embodiment For the parts not mentioned in the example, refer to the corresponding content in the method embodiment. like Figure 7 As shown, the region labeling device 300 includes:

[0148] An acquisition module 301, configured to acquire low-weight words and high-weight words in the preset data set, wherein the first weight value of the low-weight word is less than the first preset threshold, and the second weight value of the high-weight word is greater than the second preset threshold.

[0149] The filtering module 302 is configured to filter out low-weight words in the text to be marked according to the low-weight words in the preset data set.

[0150] The extraction module 3...

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 region labeling method, a device and a computer-readable storage medium, which relates to the technical field of region labeling. The method comprises the following steps of:acquiring low-weight words and high-weight words in a preset data set; extracting low-weight words conforming to a first preset threshold value and high-weight words conforming to a second preset threshold value in the preset data set as a standard for region labeling; and obtaining the low-weight words and the high-weight words in the preset data set. According to the low weight words in the preset data set; Obtaining high-weight words in the text to be annotated by filtering out low-weight high-weight words in the text to be annotated according to the high-weight words in the preset data set, and extracting regional keywords in the high-weight words; Mark the region label of the text to be marked according to the region keyword. When the method is used to extract the regional label fromthe text to be labeled of the description picture, the labeling standard of the regional label is identical, which effectively reduces the labeling error of the regional label caused by the inconsistency of the information of the text to be labeled understood by each labeling personnel, and improves the labeling precision and the labeling efficiency of the regional label.

Description

technical field [0001] The present invention relates to the technical field of regional labeling, in particular to a method, device and computer-readable storage medium for regional labeling. Background technique [0002] With the in-depth development of the Internet, the types of multimedia data are becoming more and more diverse, and the types of multimedia data include but are not limited to pictures and texts. Among them, the visual information of the picture is strong, but the semantic description is difficult, and the regional information is an important semantic description of the picture. For example, the content of some pictures can assist in analyzing the semantics of the picture through the text of the picture annotation, and is used to recommend a certain region. Gourmet merchants or places of interest, these all need to determine the geographical information of the picture. These geographical information, as the basic data of top-level services, are generally u...

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): G06F17/21G06F17/27G06F16/48
CPCG06F40/117G06F40/216G06F40/30
Inventor 王辰龙
Owner XIAMEN MEITUZHIJIA TECH
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap