CN114972346A - Stone identification method based on computer vision - Google Patents
Stone identification method based on computer vision Download PDFInfo
- Publication number
- CN114972346A CN114972346A CN202210902703.8A CN202210902703A CN114972346A CN 114972346 A CN114972346 A CN 114972346A CN 202210902703 A CN202210902703 A CN 202210902703A CN 114972346 A CN114972346 A CN 114972346A
- Authority
- CN
- China
- Prior art keywords
- image
- stone
- value
- hsi
- sliding window
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000004575 stone Substances 0.000 title claims abstract description 116
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000011156 evaluation Methods 0.000 claims abstract description 39
- 230000014509 gene expression Effects 0.000 claims description 16
- 239000000463 material Substances 0.000 claims description 4
- 238000005286 illumination Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000013527 convolutional neural network Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000005034 decoration Methods 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 235000019738 Limestone Nutrition 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000010438 granite Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 239000006028 limestone Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000011435 rock Substances 0.000 description 1
- -1 sandstone Substances 0.000 description 1
- 239000010454 slate Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G06T5/90—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
Abstract
The invention relates to the field of computer vision, in particular to a stone identification method based on computer vision, which comprises the following steps: acquiring an HSI image of a stone to be detected, extracting an I channel image in the HSI image, and acquiring a brightness characteristic value of the I channel image; obtaining the contrast of the HSI image, and obtaining the change degree of the HSI image affected by the brightness by using the contrast of the HSI image; setting a sliding window for the HSI image, obtaining a single characteristic value of each pixel point in the sliding window by using the pixel value mean value, the pixel value variance and the pixel value of the pixel point in the sliding window, obtaining the overall characteristic value of the HSI image according to the single characteristic value of each pixel point in the sliding window, further obtaining the overall evaluation value of the stone to be detected, and identifying the category of the stone to be detected by using the overall evaluation value of the stone to be detected and the overall evaluation value of each template stone. The invention improves the accuracy of stone identification.
Description
Technical Field
The invention relates to the field of computer vision, in particular to a stone identification method based on computer vision.
Background
The stone is a high-grade product of building decoration materials, natural stones are mainly divided into granite, slate, sandstone, limestone, volcanic rock and the like, along with the development of building design, the stone has already become one of important raw materials for building, decoration, road and bridge construction, different industries have different requirements on the types of the stone, the traditional method simply identifies the stone according to the brightness of the stone, but only identifies the stone by using the brightness characteristic, so that the identification result is inaccurate; the stone type identification is carried out through the CNN neural network, but the CNN neural network identification of the stone type requires a large amount of data sets for labeling, and errors are easy to occur, so that the CNN neural network identification result is inaccurate. Therefore, the invention provides a stone identification method based on computer vision, which improves the accuracy of stone identification.
Disclosure of Invention
The invention provides a stone identification method based on computer vision, which aims to solve the problem of inaccurate stone identification in the prior art.
The invention relates to a stone recognition method based on computer vision, which adopts the following technical scheme:
s1, acquiring an HSI image of the stone to be detected, extracting an I channel image in the HSI image, acquiring a gray level mean value of the I channel image and a modulus of each pixel point gradient in the I channel image, and acquiring a brightness characteristic value of the I channel image by using the gray level mean value and the modulus of each pixel point gradient in the I channel image;
s2, obtaining the contrast of the HSI image, and obtaining the change degree of the HSI image affected by the brightness by using the brightness characteristic value of the I channel image and the contrast of the HSI image;
s3, setting a sliding window, traversing in the HSI image by using the set sliding window, calculating a single characteristic value of each pixel point in the sliding window by using the pixel value of each pixel point in each sliding window, the pixel value mean value and the pixel value variance of the pixel point in the sliding window, and calculating the integral characteristic value of the sliding window by using the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points;
s4, calculating the overall characteristic value of the sliding window according to the brightness characteristic value of the I channel image, the brightness influence change degree of the HSI image, the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points to obtain the overall evaluation value of the stone to be detected, and matching the overall evaluation value of the stone to be detected with the overall evaluation value of each template stone to finish the classification of the stone to be detected.
Further, the modulus of the gradient of each pixel point is determined according to the following method:
acquiring the gradient amplitude and the gradient direction of each pixel point in the I channel image;
and obtaining a gradient vector of each pixel point by using the gradient amplitude and the gradient direction of each pixel point, and obtaining a mode of the gradient of each pixel point in the I-channel image by using the gradient vector of each pixel point.
Further, the expression of the luminance characteristic value of the I-channel image is:
in the formula:
representing the luminance characteristic value of the I-channel image,
represents the mean value of the gray levels of the I-channel image,
the number of pixel points in the I-channel image is represented,
representing the second in an I-channel image
The modulus of the gradient of the individual pixels,
representing the second in an I-channel image
And (5) each pixel point.
Further, the method for acquiring the HSI image contrast is as follows:
acquiring the relative entropy of the I channel image and the HSI image and the contrast of the I channel image;
and obtaining the contrast of the HSI image by using the ratio of the relative entropy of the I channel image and the HSI image to the contrast of the I channel image.
Further, the specific expression of the degree of change of the HSI image affected by the brightness is:
in the formula:
indicating the degree of change in the HSI image affected by brightness,
representing the luminance characteristic value of the I-channel image,
representing the contrast of the HSI image.
Further, the expression of the single feature value of each pixel point is as follows:
in the formula:
representing points in an HSI image
Is determined by the characteristic of the signal of the sensor,
indicating the length of the HSI picture,
which represents the width of the image,
representing pixel values of pixels within a sliding windowThe average value of the average value is calculated,
representing the variance of pixel values of pixels within the sliding window,
representing points in an HSI image
The pixel value of (2).
Further, the specific expression of the overall characteristic value of the HSI image is as follows:
in the formula:
represents the overall feature value of the HSI image,
is shown as
The overall characteristic value of each sliding window,
denotes the first
A sliding window is arranged on the top of the sliding window,
the number of the sliding windows is shown,
represents the maximum of the overall characteristic values of all the sliding windows,
to indicate all slipsThe minimum value among the global characteristic values of the window.
Further, the specific expression of the overall evaluation value of the stone to be detected is as follows:
in the formula:
showing the overall evaluation value of the stone to be detected, C showing the contrast of the I-channel image,
representing the influence value of the brightness and the contrast in the stone image to be detected,
a normalized numerical value representing the overall feature value of the HIS image.
Further, the method for identifying the category of the stone to be detected comprises the following steps:
obtaining the overall evaluation value of each template stone by utilizing the steps S1-S4 in the specification;
when the overall evaluation value of the stone to be detected
Overall evaluation value of form stone
Satisfy the requirement of
And meanwhile, the stone to be detected and the template stone belong to the same stone.
The invention has the beneficial effects that: the collected stone image to be detected is converted into the HIS image of the stone to be detected, the overall evaluation value of the stone is obtained by utilizing the brightness characteristic value of the I channel in the HSI image, the change degree of the HIS image influenced by the brightness and the overall characteristic value of the HIS image, and the overall evaluation value identifies the stone from three dimensions of brightness, contrast and overall information of the stone surface, so that the accuracy of stone identification is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a flowchart of a stone recognition method based on computer vision according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the stone recognition method based on computer vision of the invention, as shown in fig. 1, comprises:
s1, obtaining an HSI image of the stone to be detected, extracting an I channel image in the HSI image, obtaining a gray average value of the I channel image and a modulus of each pixel point gradient in the I channel image, and obtaining a brightness characteristic value of the I channel image by using the gray average value and the modulus of each pixel point gradient in the I channel image.
The method comprises the steps of collecting an image of the surface of a stone to be detected, wherein in the process of collecting the image, the surface of the stone can be reflected by light, so that the illumination is uniform when the image is collected. In order to make the illumination uniform, the illumination direction of the light is set to be overlook illumination, and a plurality of light sources are needed, so that uniform light is formed and is irradiated on the surface of the stone to obtain the stone image to be detected. And the flawless stone images of all models produced and processed by factories are required to be collected to be used as template stone images, the characteristic parameters of all the stone images to be detected and the template stone images are calculated, and the categories of the stones to be detected are identified according to the parameters.
The captured RGB image is converted into an HSI image because in an HSI image, H is called hue to define the wavelength of the color, S is called saturation to define the shade of the color, and I is used to define the brightness. The HSI image can represent the color information characteristic of one image, so the invention calculates according to the HIS image and respectively obtains the overall evaluation values of the stone image to be detected and the template stone image.
The method comprises the steps of extracting an I channel image in a stone material HSI image, obtaining the mean value of all pixel point gray values in the I channel image, taking the mean value of all pixel point gray values in the I channel image as the gray mean value of the I channel image, obtaining the gradient amplitude and the gradient direction of each pixel point in the I channel image, obtaining the gradient vector of each pixel point by using the gradient amplitude and the gradient direction of each pixel point in the I channel image, obtaining the mode of each pixel point gradient in the I channel image by using the gradient vector of each pixel point, obtaining the brightness characteristic value of the I channel image by using the gray mean value and the mode of each pixel point gradient in the I channel image, and adopting the following specific expressions:
in the formula:
representing the luminance characteristic value of the I-channel image,
represents the mean value of the gray levels of the I-channel image,
the number of pixel points in the I-channel image is represented,
representing the second in an I-channel image
The modulus of the gradient of the individual pixels,
representing the second in an I-channel image
And (5) each pixel point.
Thus, the brightness characteristic value of the I-channel image is obtained.
And S2, acquiring the contrast of the HSI image, and obtaining the change degree of the HSI image influenced by the brightness by using the brightness characteristic value of the I channel image and the contrast of the HSI image.
Entropy of an image is a statistical form of features that reveal how much information is averaged out in the image. The greater the amount of information contained in an image, the greater the entropy of the image and the greater the contrast of the image. Because the image has different distortions in multiple channels, the contrast characteristic of a single channel is calculated, and the distortion of the image changes, so the relative entropy of the image is used for measuring the distortion degree of the single-channel image and the multi-channel image.
The method comprises the following specific steps of obtaining the relative entropy of an I channel image and an HSI image: respectively acquiring the number of pixel points in an I channel image and an HSI image; acquiring the number of pixel points corresponding to each same pixel value in the I channel image and the HSI image; calculating the ratio of the number of pixel points corresponding to each same pixel value in the I channel image and the HSI image to the total number of the pixel points in the I channel image; calculating the ratio of the number of pixel points corresponding to each same pixel value in the I channel image and the HSI image to the total number of pixel points in the HSI image; obtaining the relative entropy of the I channel image and the HSI image by utilizing the ratio of the number of the pixel points corresponding to each same pixel value in the I channel image and the HSI image to the number of the pixel points in the I channel image and the ratio of the number of the pixel points corresponding to each same pixel value in the I channel image and the HSI image to the number of the pixel points in the HSI image, wherein the specific expression of the relative entropy of the I channel image and the HSI image is as follows:
in the formula:
representing the relative entropy of the I channel image and the HSI image,
the ratio of the number of the pixel points corresponding to each same pixel value in the I channel image and the HSI image to the number of the pixel points in the I channel image is represented,
the ratio of the number of the pixel points corresponding to each same pixel value in the I channel image and the HSI image to the number of the pixel points in the HSI image is represented,
indicating that the same pixel value in the I-channel image and the HSI image is
The number of the pixels in time is,
indicating the number of identical pixel values in the same I channel image and HSI image.
Wherein when
And
the higher the degree of similarity is, the higher,
the smaller the value, the smaller the relative entropy, indicating that the less the I-channel image contrast distortion, the higher the perceived quality of the image. Otherwise, the I channel image is comparedThe greater the degree distortion, the lower the image perceived quality. The relative entropy is used as a contrast characteristic to measure the distortion of the image.
Obtaining contrast of I-channel images
Using the relative entropy of the I-channel image and the HSI image
Contrast of I-channel image
Ratio of
Obtaining contrast of HSI image
. Contrast ratio according to HSI image
And obtaining the variation degree of the HSI image influenced by the brightness according to the brightness characteristic value of the I channel image
The specific expression is as follows:
in the formula:
indicating the degree of change in the HSI image affected by brightness,
representing the luminance characteristic value of the I-channel image,
representing the contrast of the HSI image.
Wherein, the contrast ratio normalization value of the HSI image is subtracted by the whole brightness normalization value of the brightness channel to represent the change coefficient of the contrast ratio, and the characteristic quantity is used for evaluating the change degree of the HSI image influenced by the brightness.
Thus, the degree of change of the HSI image affected by the brightness is obtained.
S3, setting a sliding window, traversing in the HSI image by using the set sliding window, calculating a single characteristic value of each pixel point in the sliding window by using the pixel value of each pixel point in each sliding window, the pixel value mean value and the pixel value variance of the pixel point in the sliding window, and calculating the overall characteristic value of the sliding window by using the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points.
And calculating the overall characteristic value of the HSI image according to the pixel values of the pixel points in the HSI image, wherein the characteristic is used for evaluating the HSI image quality. Because the texture distribution on the surface of the stone image is irregular, and the stone image is matched according to all parameters of the image in the process of identifying and matching the stone image, the overall characteristic value of the HSI image is calculated to judge whether the overall characteristic value of the stone image to be detected is consistent with that of the template image.
The method comprises the following steps of obtaining a single characteristic value of each pixel point in a sliding window: setting 3
Window of 3 size, will 3
A window of size 3 traverses the HSI image as a sliding window. The method comprises the steps of obtaining pixel values of pixel points in a sliding window, calculating the mean value and variance of the pixel values of the pixel points in the sliding window, and obtaining a single characteristic value of each pixel point in the sliding window by using the pixel value of each pixel point in the sliding window, the mean value and variance of the pixel values of the pixel points in the sliding window, wherein the specific expression is as follows:
in the formula:
representing points in an HSI image
Is determined by the characteristic of the signal of the sensor,
indicating the length of the HSI picture,
the width of the representation image is shown,
represents the mean value of pixel values of pixel points in the sliding window,
representing the variance of pixel values of pixels within the sliding window,
representing points in an HSI image
The pixel value of (2).
In order to describe the relevance between adjacent image points in an HSI image, the relevance between a central pixel point and a neighborhood pixel point in a sliding window is obtained by utilizing a single characteristic value of each pixel point in the HSI image, the relevance between the central pixel point and the neighborhood pixel point in the sliding window is obtained by multiplying the central pixel point in the sliding window with the neighborhood pixel point respectively, the relevance between the central pixel point and each neighborhood pixel point in the sliding window is weighted and averaged to obtain the weighted average value of each sliding window, and the weighted average value of the sliding window is used as the integral characteristic value of each sliding window
。
When the relevance of the central pixel point and each neighborhood pixel point is weighted, the occupied weight is the same, and the weighted average value of the sliding window is used as the integral characteristic value of each sliding window.
Obtaining the integral characteristic value of the HSI image by utilizing the integral characteristic value of the sliding window, wherein the specific expression is as follows:
in the formula:
represents the overall feature value of the HSI image,
is shown as
The overall characteristic value of each sliding window,
is shown as
A sliding window is arranged on the top of the sliding window,
the number of the sliding windows is shown,
represents the maximum of the overall characteristic values of all the sliding windows,
represents the minimum of the overall characteristic values of all sliding windows.
Thus, the overall characteristic value of the HIS image is obtained.
S4, calculating the overall characteristic value of the sliding window according to the brightness characteristic value of the I channel image, the brightness influence change degree of the HSI image, the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points to obtain the overall evaluation value of the stone to be detected, and matching the overall evaluation value of the stone to be detected with the overall evaluation value of each template stone to finish the classification of the stone to be detected.
Obtaining an overall evaluation value of the stone to be detected according to the brightness characteristic value of the I channel image, the change degree of the HSI image affected by the brightness and the overall characteristic value of the HSI image, wherein the specific expression is as follows:
in the formula:
showing the overall evaluation value of the stone to be detected,
representing the influence value of the brightness and the contrast in the stone image to be detected,
a normalized numerical value representing the overall feature value of the HIS image.
The integral evaluation value formula of the stone to be detected represents the integral evaluation value of the stone to be detected according to the brightness characteristic value of an I channel image in an HIS image of the stone to be detected, the influence degree of the brightness characteristic value on the contrast of the HIS image, namely the influence degree of the brightness of the HIS image and the normalized numerical value of the integral characteristic value of the HIS image. Therefore, the overall characteristics of the stone image can be reflected according to the brightness characteristic value of the I channel image in the HIS image of the stone to be detected, the influence degree value of the contrast of the HIS image, namely the influence degree of the HIS image on the brightness and the overall characteristic value of the HIS image.
Using exponential functions
For amplifying the I-channel luminance characteristic value
I channel contrast ratio
HIS image influenced by brightness
，
A normalized numerical value representing the overall characteristic value of the HIS image,
when the stone characteristics to be detected are not obvious, the overall evaluation value of the stone image to be detected is amplified, so that the identification is more accurate.
Thus, the overall evaluation value of the stone to be detected is obtained
。
Using the overall evaluation value of the stone to be detected
The overall evaluation value of each template stone material
Comparing, and when the overall evaluation value of the stone to be detected
Overall evaluation value of form stone
Satisfy the requirement of
Then, the stone to be detected and the template stone belong to the same stone; overall evaluation value of each template stone
The obtaining mode is the same as the obtaining method of the overall evaluation value of the stone to be detected, and the invention is not described in detail.
The invention has the beneficial effects that: the collected stone image to be detected is converted into the HIS image of the stone to be detected, the overall evaluation value of the stone is obtained by utilizing the brightness characteristic value of the I channel in the HSI image, the change degree of the HIS image influenced by the brightness and the overall characteristic value of the HIS image, and the overall evaluation value identifies the stone from three dimensions of brightness, contrast and overall information of the stone surface, so that the accuracy of stone identification is improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. A stone recognition method based on computer vision is characterized by comprising the following steps:
s1, acquiring an HSI image of the stone to be detected, extracting an I channel image in the HSI image, acquiring a gray level mean value of the I channel image and a modulus of each pixel point gradient in the I channel image, and acquiring a brightness characteristic value of the I channel image by using the gray level mean value and the modulus of each pixel point gradient in the I channel image;
s2, obtaining the contrast of the HSI image, and obtaining the change degree of the HSI image affected by the brightness by using the brightness characteristic value of the I channel image and the contrast of the HSI image;
s3, setting a sliding window, traversing in the HSI image by using the set sliding window, calculating a single characteristic value of each pixel point in the sliding window by using the pixel value of each pixel point in each sliding window, the pixel value mean value and the pixel value variance of the pixel point in the sliding window, and calculating the integral characteristic value of the sliding window by using the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points;
s4, calculating the integral characteristic value of the sliding window according to the brightness characteristic value of the I channel image, the change degree of the HSI image influenced by the brightness, the single characteristic value of the central pixel point in each sliding window and the single characteristic values of other pixel points to obtain the integral evaluation value of the stone to be detected, and matching the integral evaluation value of the stone to be detected with the integral evaluation value of each template stone to finish the classification of the stone to be detected.
2. A method as claimed in claim 1, wherein the mode of each pixel gradient is determined as follows:
acquiring the gradient amplitude and the gradient direction of each pixel point in the I channel image;
and obtaining a gradient vector of each pixel point by using the gradient amplitude and the gradient direction of each pixel point, and obtaining a mode of the gradient of each pixel point in the I-channel image by using the gradient vector of each pixel point.
3. The computer vision-based stone recognition method as claimed in claim 1, wherein the expression of the luminance characteristic value of the I-channel image is:
in the formula:
representing the luminance characteristic value of the I-channel image,
represents the mean value of the gray levels of the I-channel image,
the number of pixel points in the I-channel image is represented,
representing the second in an I-channel image
The modulus of the gradient of the individual pixels,
representing the second in an I-channel image
And (5) each pixel point.
4. A method for stone recognition based on computer vision as claimed in claim 1, wherein said method for obtaining HSI image contrast is:
acquiring the relative entropy of the I channel image and the HSI image and the contrast of the I channel image;
and obtaining the contrast of the HSI image by using the ratio of the relative entropy of the I channel image and the HSI image to the contrast of the I channel image.
5. A stone recognition method based on computer vision as claimed in claim 1, wherein the specific expression of the degree of change of the HSI image affected by brightness is:
in the formula:
indicating the degree of change in the HSI image affected by brightness,
representing the luminance characteristic value of the I-channel image,
representing the contrast of the HSI image.
6. A method as claimed in claim 1, wherein the expression of the single eigenvalue of each pixel point is:
in the formula:
representing points in an HSI image
Is determined by the characteristic value of the single feature,
indicating the length of the HSI picture,
which represents the width of the image,
represents the mean value of pixel values of pixel points in the sliding window,
representing the variance of pixel values of pixels within the sliding window,
representing points in an HSI image
The pixel value of (2).
7. A stone recognition method based on computer vision as claimed in claim 1, wherein the specific expression of the overall feature value of the HSI image is:
in the formula:
represents the overall feature value of the HSI image,
is shown as
The overall characteristic value of each sliding window,
is shown as
A sliding window is arranged on the top of the sliding window,
the number of the sliding windows is shown,
represents the maximum of the overall characteristic values of all the sliding windows,
represents the minimum of the overall characteristic values of all sliding windows.
8. The computer vision-based stone recognition method as claimed in claim 1, wherein the specific expression of the overall evaluation value of the stone to be detected is:
in the formula:
showing the overall evaluation value of the stone to be detected, C showing the contrast of the I-channel image,
representing the influence value of the brightness and the contrast in the stone image to be detected,
a normalized numerical value representing the overall feature value of the HIS image.
9. A stone recognition method based on computer vision as claimed in claim 1, characterized in that said method of recognizing the category of stone to be detected is:
obtaining the overall evaluation value of each template stone according to the steps S1-S4;
when the overall evaluation value of the stone to be detected
Evaluation value of the whole of the stone material and the template
Satisfy the requirement of
And meanwhile, the stone to be detected and the template stone belong to the same stone.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210902703.8A CN114972346B (en) | 2022-07-29 | 2022-07-29 | Stone identification method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210902703.8A CN114972346B (en) | 2022-07-29 | 2022-07-29 | Stone identification method based on computer vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114972346A true CN114972346A (en) | 2022-08-30 |
| CN114972346B CN114972346B (en) | 2022-11-04 |
Family
ID=82968965
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210902703.8A Active CN114972346B (en) | 2022-07-29 | 2022-07-29 | Stone identification method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114972346B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164737A (en) * | 2005-12-16 | 2007-06-28 | Mitsubishi Heavy Ind Ltd | Color-identifying method |
| US20150227816A1 (en) * | 2014-02-10 | 2015-08-13 | Huawei Technologies Co., Ltd. | Method and apparatus for detecting salient region of image |
| WO2017177717A1 (en) * | 2016-04-14 | 2017-10-19 | 广州视源电子科技股份有限公司 | Element positioning method and system based on color and gradient |
| CN109961415A (en) * | 2019-03-26 | 2019-07-02 | 常州工学院 | A kind of adaptive gain underwater picture Enhancement Method based on HSI space optics imaging model |
| CN110807817A (en) * | 2019-10-29 | 2020-02-18 | 长春融成智能设备制造股份有限公司 | Machine vision method for target color recognition adapting to illumination change |
| CN111626967A (en) * | 2019-02-28 | 2020-09-04 | 富华科精密工业（深圳）有限公司 | Image enhancement method, image enhancement device, computer device and readable storage medium |
| CN111861937A (en) * | 2020-07-29 | 2020-10-30 | 西安建筑科技大学 | MSR (minimum shift register) improvement-based image enhancement method and system |
| CN113989275A (en) * | 2021-12-10 | 2022-01-28 | 沭阳县源美装饰材料有限公司 | Initial weight optimization-based wood board type identification method and device of neural network |
| CN114170208A (en) * | 2021-12-14 | 2022-03-11 | 武汉福旺家包装有限公司 | Paper product defect detection method based on artificial intelligence |
| CN114445387A (en) * | 2022-01-29 | 2022-05-06 | 泗阳富艺木业股份有限公司 | Fiberboard quality classification method based on machine vision |
-
2022
- 2022-07-29 CN CN202210902703.8A patent/CN114972346B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007164737A (en) * | 2005-12-16 | 2007-06-28 | Mitsubishi Heavy Ind Ltd | Color-identifying method |
| US20150227816A1 (en) * | 2014-02-10 | 2015-08-13 | Huawei Technologies Co., Ltd. | Method and apparatus for detecting salient region of image |
| WO2017177717A1 (en) * | 2016-04-14 | 2017-10-19 | 广州视源电子科技股份有限公司 | Element positioning method and system based on color and gradient |
| CN111626967A (en) * | 2019-02-28 | 2020-09-04 | 富华科精密工业（深圳）有限公司 | Image enhancement method, image enhancement device, computer device and readable storage medium |
| CN109961415A (en) * | 2019-03-26 | 2019-07-02 | 常州工学院 | A kind of adaptive gain underwater picture Enhancement Method based on HSI space optics imaging model |
| CN110807817A (en) * | 2019-10-29 | 2020-02-18 | 长春融成智能设备制造股份有限公司 | Machine vision method for target color recognition adapting to illumination change |
| CN111861937A (en) * | 2020-07-29 | 2020-10-30 | 西安建筑科技大学 | MSR (minimum shift register) improvement-based image enhancement method and system |
| CN113989275A (en) * | 2021-12-10 | 2022-01-28 | 沭阳县源美装饰材料有限公司 | Initial weight optimization-based wood board type identification method and device of neural network |
| CN114170208A (en) * | 2021-12-14 | 2022-03-11 | 武汉福旺家包装有限公司 | Paper product defect detection method based on artificial intelligence |
| CN114445387A (en) * | 2022-01-29 | 2022-05-06 | 泗阳富艺木业股份有限公司 | Fiberboard quality classification method based on machine vision |
Non-Patent Citations (4)
| Title |
|---|
| ANYONG QIN, ET AL: "Edge-Smoothing-Based Distribution Preserving Hyperspherical Embedding for Hyperspectral Image Classification", 《IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING》 * |
| 侯之旭等: "一种彩色图像分割的障碍物识别方法", 《重庆理工大学学报(自然科学)》 * |
| 王晓等: "基于多特征融合的彩色图像声呐目标检测", 《计算机科学》 * |
| 陈先桥等: "一种考虑分类的石材图像相似性比对算法", 《武汉理工大学学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114972346B (en) | 2022-11-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109269951B (en) | Image-based flotation tailing ash content, concentration and coarse particle content detection method | |
| CN110097034B (en) | Intelligent face health degree identification and evaluation method | |
| CN100550037C (en) | Utilize and improve Hausdorff apart from the method for extracting the identification human ear characteristic | |
| CN111968080B (en) | Method for detecting quality of inside and outside of Feicheng peaches based on hyperspectral and deep learning | |
| CN103518224B (en) | Method for analysing microbial growth | |
| CN113554629A (en) | Strip steel red rust defect detection method based on artificial intelligence | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN110781889B (en) | Deep learning-based nondestructive testing method for total sugar content in blueberry fruits | |
| CN109284718B (en) | Inspection robot-oriented variable-view-angle multi-instrument simultaneous identification method | |
| CN112215790A (en) | KI67 index analysis method based on deep learning | |
| CN111652883B (en) | Glass surface defect detection method based on deep learning | |
| CN110782440B (en) | Crop seed character measuring method | |
| CN107610119B (en) | The accurate detection method of steel strip surface defect decomposed based on histogram | |
| CN116468724B (en) | Method for testing content of calcium carbonate based on optical sensing technology | |
| CN115994907B (en) | Intelligent processing system and method for comprehensive information of food detection mechanism | |
| CN111160451A (en) | Flexible material detection method and storage medium thereof | |
| CN116580021A (en) | Environment-friendly concrete carbon reduction product production and quality detection method | |
| CN113724339B (en) | Color space feature-based color separation method for tiles with few samples | |
| JP2020169994A (en) | Method for optically detecting biomarkers | |
| CN109816629B (en) | Method and device for separating moss based on k-means clustering | |
| CN114972346B (en) | Stone identification method based on computer vision | |
| JP2840347B2 (en) | Board mounting inspection equipment | |
| CN116246174A (en) | Sweet potato variety identification method based on image processing | |
| CN116109577A (en) | Printing label defect detection system and method | |
| CN111563536B (en) | Bamboo strip color self-adaptive classification method based on machine learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |