A new threshold segmentation method of color invariance of colored image
A color image and threshold segmentation technology, applied in the field of image processing, can solve the problems of image color recognition interference, affect the accuracy of image segmentation, etc., and achieve good adaptability
Inactive Publication Date: 2007-05-23
SHANDONG UNIV
View PDF0 Cites 6 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The purpose of the present invention is to solve the current problems that cannot effectively solve the interference of noise and light factors on image color recognition, affecting the accuracy of image segmentation, etc., and provide a device that can effectively eliminate the diffuse reflection surface and specular reflection in the environment of point light sources. The influence of the surface on the precise threshold segmentation, and a new color image color invariance threshold segmentation method that satisfies the advantages of threshold segmentation accuracy and robustness in the point light source environment
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0022] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0023] In Fig. 1, an image acquisition device 2 (such as a digital camera) is used to acquire an image of the surface of the object, and at least one point light source 1 and its normal projection point P at any position on the image are selected. s and several arbitrary reference points P that do not overlap with the projection point i ; h1 and h2 are the distances from the light source and the camera to the plane, respectively.
[0024] The Phong lighting model used is I λ = I Lλ cos θ + I Sλ cos n λ α , where I λ is reflected light, I Lλ is the diffuse reflection coefficient, I Sλ is the specular reflection coeffici...
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
Login to View More Abstract
The invention discloses a new multicolor image color invariant threshold value division method. It solves the currently not-effective problem that the image color recognition interference, such as the noise and light factors, affects the accuracy of image division. It can effective eliminate the influence of threshold value accuracy by the diffuse reflectance surface and the mirror reflection surface in the point light source environment, and has the advantages of meeting accuracy and robustness of threshold value division in the point light source environment. Its method provides a new measurement of the reflectivity of the diffuse reflectance surface and the mirror reflection surface through deriving the color invariance of Phong illumination model, and using Gaussian colors similar model to achieve an adaptive threshold value method of the target pixel-level.
Description
technical field [0001] The invention relates to an image processing method, in particular to a new color image color invariance threshold segmentation method. Background technique [0002] The purpose of image threshold segmentation is to separate the background and target pixels in the image. After image thresholding, object pixels can be represented by a different color than background pixels. The ideal situation is that all object pixels are completely selected, but due to the influence of noise and illumination, it is usually difficult to find a single threshold method that can be effective for any color image, even if a single threshold is used for an image. ideal. Therefore, there are three main methods for eliminating the influence of illumination at present: one is the gradient method, the other is the color statistical prediction method, and the other is the statistical prediction method based on the two-tone model. The first two algorithms work well for the iden...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/34G06V10/60
CPCG06K9/4661G06V10/60
Inventor 蔡珣孟祥旭蔡菲
Owner SHANDONG UNIV