Image analysis device, image analysis system, and image analysis method

An image analysis, image technology, applied in image data processing, image data processing, instruments, etc., can solve the problem of unrealistic view of processing time

Active Publication Date: 2017-04-26
HITACHI LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0015] The technique described in Non-Patent Document 1 is effective when, for example, the detection target is determined in advance, such as face detection. However, in applications where the user specifies the detection target one by one, it is necessary to collect samples and mechanically perform each time. learning, so not realistic from a processing time point of view

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
  • Image analysis device, image analysis system, and image analysis method
  • Image analysis device, image analysis system, and image analysis method
  • Image analysis device, image analysis system, and image analysis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0038]

[0039] figure 1 It is a configuration diagram of the image analysis system 100 according to Embodiment 1 of the present invention. The purpose of the image analysis system 100 is to search for an image containing an arbitrary object specified by a user from a large number of images. The image analysis system 100 includes: an image / text storage device 101 , an input device 102 , a display device 103 , a data storage device 104 , an image database 105 , and an image analysis device 106 .

[0040] The image / text storage device 101 is a storage medium for storing image data, and can be configured using a hard disk drive attached to a computer, NAS (Network Attached Storage), or a storage system connected through a network such as SAN (Storage Area Network). The scale of the image data to be analyzed by the image analysis system 100 is assumed to be, for example, a large scale of hundreds of thousands of pieces.

[0041] The input device 102 is an input interface for t...

Embodiment approach 2

[0151] In the image analysis system 100 described in Embodiment 1, the analysis target determination unit 109 narrows down images to be detected as objects by searching bibliographic information. Therefore, even if an object specified by the user is actually included, an image with insufficient bibliographic information is not subject to detection processing and does not appear as an analysis result. A method of reducing omission of detection processing objects by expanding bibliographic information will be described below. The other configurations are substantially the same as those in Embodiment 1, so the differences will be explained below with emphasis.

[0152] Figure 11 It is a figure explaining the example of extended bibliography information. For comparison, in Figure 11 (a) shows a search concept diagram of the case where the bibliographic information is not expanded, Figure 11 (b) A search conceptual diagram showing a case where bibliographic information is ex...

Embodiment approach 3

[0178] In Embodiment 3 of the present invention, a method of improving the accuracy of object detection by using intermediate data in processing of the image analysis system 100 will be described. This method is used as an object detection method using multiple Figure 7 Description template. Other configurations are the same as those in Embodiments 1 and 2, so the following description focuses on the point of using multiple templates to search for similar images when generating metadata of a query image.

[0179] Figure 15 It is a figure explaining the method of improving the accuracy of an object detection by expanding the template used when searching for the image similar to a query image. Use the following Figure 15 Expansion of the template in Embodiment 3 will be described.

[0180] in use Figure 7 In the described object detection method, an object region is specified by examining a partial region of an image, a template, and a degree of similarity. Therefore, ...

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

An object of the present invention is to provide an image analysis technique capable of detecting a detection target object from image data at high speed. The image analysis device of the present invention generates metadata including a query image of an object to be detected, uses the metadata to narrow down image data to be searched in advance, and then performs object detection.

Description

technical field [0001] The present invention relates to techniques for detecting specific objects contained in image data. Background technique [0002] With the development of IT infrastructure for individuals / enterprises, a large amount of multimedia data (text, video / image, sound, various log data, etc.) is stored in large-scale storage. In order to efficiently find information from a large amount of stored data, various information retrieval techniques have been considered for each piece of media data and have been gradually put into practical use. [0003] As an example of information retrieval for multimedia data, it is conceivable to detect objects and specific regions contained in images. Object detection and area determination in images is equivalent to morpheme analysis in text analysis (a method of dividing text into words and determining parts of speech), and is an important key technology when analyzing the meaning of images. [0004] The method of Non-Patent ...

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 Patents(China)
IPC IPC(8): G06T1/00G06F17/30
CPCG06F16/583G06F16/5854G06V30/242G06V2201/10G06V2201/07G06F18/211G06F18/214
Inventor 渡边裕树广池敦
Owner HITACHI LTD
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