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

Improved naive Bayes-based indoor visual feature classification method

A technology of visual features and classification methods, applied in special data processing applications, instruments, electrical and digital data processing, etc. Effect

Active Publication Date: 2018-07-06
HARBIN VOCATIONAL & TECHN COLLEGE
View PDF6 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the problems of poor matching accuracy and slow matching speed of traditional indoor visual feature classification methods, the present invention proposes an indoor visual feature classification method based on improved naive Bayesian

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
  • Improved naive Bayes-based indoor visual feature classification method
  • Improved naive Bayes-based indoor visual feature classification method
  • Improved naive Bayes-based indoor visual feature classification method

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0021] Specific implementation mode one: as figure 1 As shown, an indoor visual feature classification method based on improved naive Bayesian given in this embodiment is implemented according to the following steps:

[0022] Step 1: Applying the SURF algorithm to extract features from the video database image;

[0023] Step 2: Apply an iterative comparison algorithm to classify the SURF features of the video database image extracted in step 1, and generate a SURF feature tree;

[0024] Step 3: Apply a dimension selection algorithm to select the dimension of the SURF feature of the video database image whose variance of the mean value of the dimension exceeds the threshold;

[0025] Step 4: Generate an improved Naive Bayesian algorithm model according to the dimensions selected in Step 3;

[0026] Step 5: Use the SURF algorithm to extract SURF features from the user positioning picture;

[0027] Step 6: Input the extracted SURF feature of the user positioning picture into t...

specific Embodiment approach 2

[0029] Specific embodiment two: the difference between this embodiment and specific embodiment one is: in step one, the process of applying the SURF algorithm to the video database image for feature extraction includes:

[0030] Step 11. Feature point detection:

[0031] Construct a scale space, convolve box filters of different scales with video database images and construct a scale space pyramid to form a multi-scale space function D xx ,D yy ,D xy ; where D xx Representing points on video database images with Gaussian second order partial derivatives The result of convolution, where D yy Representing points on video database images with Gaussian second order partial derivatives The result of convolution, where D xy Representing points on video database images with Gaussian second order partial derivatives The result of convolution; x represents the abscissa of the point on the video database image, y represents the ordinate of the point on the video database image...

specific Embodiment approach 3

[0039] Embodiment 3: The difference between this embodiment and Embodiment 2 is that in step 5, the process of extracting SURF features from user positioning pictures is the same as the method of using SURF algorithm to extract features from video database images in step 1.

[0040] Other steps and parameters are the same as those in Embodiment 1 or 2.

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 relates to an indoor visual feature classification method, in particular, an improved naive Bayes-based indoor visual feature classification method and belongs to the indoor positioningimage processing technical field. The method of the invention includes the following steps that: an SURF algorithm is adopted to perform feature extraction on video database images; an iterative comparison algorithm is adopted to classify extracted features, so that an SURF feature tree is generated; a dimensionality selection algorithm is adopted to select the dimensionalities of the SURF features of the video database images, wherein the variance of the mean value of the dimensionalities exceeds a threshold value; and an improved naive Bayesian algorithm model is generated; SURF feature extraction is performed on a user location image; the extracted features of the user location image are inputted into the improved naive Bayesian classification algorithm model, so that a category to which the SURF features of the user location image belong is obtained. With the method of the invention adopted, the problems of poor matching precision and slow matching speed of a traditional indoor visual feature classification method can be solved. The method of the invention can be applied to an indoor visual positioning system.

Description

technical field [0001] The invention belongs to the technical field of indoor positioning image processing, and in particular relates to an indoor feature classification method. Background technique [0002] In the field of visual positioning in image processing technology, visual positioning needs to use rich image information to complete the positioning work, and any type of visual indoor positioning method involves an accurate positioning process. The existing precise positioning process of indoor visual positioning generally adopts the method of comparing the Euclidean distance of the feature vector based on the reference frame, but this method has large errors and poor matching accuracy due to the lack of an effective reference frame selection strategy; The feature classification method of the Adams classifier can select a reference frame with a small error, but the positioning effect is unstable, and the classification time depends on the number of sample classificatio...

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): G06K9/62G06K9/48G06K9/46G06K9/00G06F17/30
CPCG06F16/783G06V20/36G06V20/46G06V10/462G06V10/46G06V10/478G06F18/24155
Inventor 殷锡亮郭娜朱娜
Owner HARBIN VOCATIONAL & TECHN COLLEGE