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

Feature extraction device, feature extraction program, and image processing device

A feature extraction, pixel technology, applied in image communication, instrument, character and pattern recognition, etc., to achieve the effect of suppressing the increase in processing load

Active Publication Date: 2017-05-17
PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, depending on the type of image or the type of object to be detected, the area to be calculated is set wider, and the local binary pattern is generated from more pixels, so there is a commonality of wanting to use a wider range of features. current requirements

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
  • Feature extraction device, feature extraction program, and image processing device
  • Feature extraction device, feature extraction program, and image processing device
  • Feature extraction device, feature extraction program, and image processing device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0067] Embodiment 1 of the present invention is an example of the basic form of the present invention.

[0068] figure 1 It is a block diagram showing an example of the configuration of the feature extraction device according to this embodiment.

[0069] exist figure 1 Among them, the feature extraction device 10 has a sub-region setting unit 433 and a binary pattern generating unit 434 .

[0070] The sub-region setting unit 433 sets a plurality of sub-regions for each of all or some pixels of the image, using the pixel as a pixel of interest. At this time, sub-region setting section 433 sets at least a region composed of a plurality of pixels including pixels separated from the pixel of interest as a sub-region.

[0071] Binary pattern generation section 434 generates, for each pixel of interest, a local binary pattern expressed by a bit value and a pixel value comparison between each of the set plurality of subregions. At this time, binary pattern generating section 434 ...

Embodiment approach 2

[0075] Embodiment 2 of the present invention is an example of a specific form in which the present invention is applied to an object detection system.

[0076] First, the configuration of an object detection system including the feature extraction device of this embodiment will be described.

[0077] figure 2 It is a block diagram showing an example of the configuration of an object detection system including the feature extraction device of this embodiment.

[0078] exist figure 2 Among them, the object detection system 100 includes a classifier learning device 200 , a classifier storage device 300 , and an object detection device 400 . The classifier learning device 200 and the object detection device 400 can be respectively connected to the classifier storage device 300 via a communication network such as the Internet, for example.

[0079] In this embodiment, as an example of the machine learning method used by the classifier learning device 200 , a Boosting method is...

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 present invention provides a feature extraction device capable of generating a local binary pattern from more pixels while suppressing a decrease in object detection accuracy and suppressing an increase in processing load. This feature extraction device includes a sub-region setting unit (433) that sets a plurality of sub-regions for a pixel of interest, and generates a binary pattern for each pixel of interest that represents a local binary pattern for comparison with pixel values ​​between the sub-regions. A generation unit (434), a sub-region setting unit (433) at least sets an area composed of a plurality of pixels as a sub-region and the plurality of pixels includes pixels away from the pixel of interest, and a binary pattern generation unit (434) for each A representative value is calculated for each sub-region (437), and a local binary pattern (439) indicating whether the difference (438) between the representative value and the pixel value of the pixel of interest is above a predetermined threshold is generated (439).

Description

technical field [0001] The present invention relates to a feature extraction device for extracting image features from image data, a feature extraction program, and an image processing device using the feature extraction device. Background technique [0002] Conventionally, a process of extracting image features from image data to detect or recognize objects contained in the image (hereinafter referred to as "object detection") has been widely performed. As one method of object detection, for example, non-patent document 1 describes a technique using a local binary pattern (LBP). [0003] The local binary pattern is a binary pattern obtained by binarizing and arranging the differences in pixel values ​​between each pixel of interest and pixels in the vicinity of the pixel of interest. According to the local binary mode, the grayscale mode contained in the image can be extracted. [0004] In the techniques described in Non-Patent Document 1 and Non-Patent Document 2 (herein...

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): G06K9/46G06V10/50
CPCG06V10/467G06V10/50H04N21/4728H04N2201/33371
Inventor 曹芸芸西村洋文S.普拉纳塔Z.H.牛
Owner PANASONIC INTELLECTUAL PROPERTY MANAGEMENT CO LTD