CN114549808A - Method for extracting image characteristics of roof in coal roadway empty-roof area - Google Patents

Abstract

The invention provides a method for extracting image characteristics of a top plate of a coal roadway empty roof area, which comprises the following steps: s1, after the fully mechanized coal cutter cuts the coal body, adjusting the position of a camera lens to align to the exposed top plate of the empty roof area; s2, collecting a naked top image by using an aboveground connection computer; s3, preprocessing the bare top image, and respectively adjusting and eliminating the brightness and noise of the image by gamma correction and median filtering; s4, extracting the contour of the rock mass structural plane from the picture through region growing segmentation and Hough transformation; s5, thinning the structural surface outline and connecting the interval points to obtain a structural surface framework; s6, calculating structural surface characteristic parameters including crack gap width and roughness, and classifying the roadway bare top; the method for extracting the roof image characteristics of the empty roof area is fully-automatic, manual intervention is not required in the whole process, the qualitative evaluation is accurately carried out on the image of the empty roof, and a better data basis is provided for intelligent tunneling advanced detection and how to support and intervene.

Description

A method for extracting image features of roof in empty roof area of coal roadway

technical field

The present application relates to the technical field of open roofs of coal roadways, and in particular, to a method for extracting image features of roofs in open roof areas of coal roadways.

Background technique

The intelligence of the coal industry is an inevitable requirement of the times, and intelligent excavation is an important part of the intelligentization of coal mines. For the intelligent excavation of coal roadway, it is necessary to emphasize the control of the roof of the coal roadway, which is related to whether the intelligent excavation can be carried out safely and efficiently continuously.

The roadway support always lags behind the rock breaking of the fully mechanized excavator for a certain distance, which inevitably results in a certain range of exposed space. This part of the roof is called the empty roof area. The empty roof area has a certain self-stabilizing performance, but there is currently no qualitative Therefore, it is impossible to provide advance prediction for intelligent excavation, so a method for extracting the roof image features of the empty roof area of coal roadway is urgently needed.

SUMMARY OF THE INVENTION

In order to make up for the above deficiencies, the present application provides a method for extracting the image features of the roof of the empty roof area of a coal roadway, aiming to improve the current lack of a qualitative standard to determine the self-stabilization performance of the roof area, so that it cannot provide advanced technology for intelligent excavation. Anticipate.

The embodiment of the present application provides a method for extracting image features of a roof in a hollow roof area of a coal roadway, including the following method steps:

S1. After the fully mechanized excavator finishes cutting the coal body, adjust the position of the camera lens to aim at the exposed roof of the empty roof area;

S2. Use the well-connected computer to collect bare-top images;

S3. Preprocess the bare-top image, and use gamma correction and median filtering to adjust and eliminate the brightness and noise of the image;

S4. The image is extracted by region growth segmentation and Hough transform to extract the contour of rock mass structure surface;

S5. Refine the outline of the structural surface and connect the space points to obtain the skeleton of the structural surface;

S6. Calculate the characteristic parameters of the structure surface, including the width and roughness of the crack gap, and classify the bare roof of the roadway.

In the above implementation process, the method can be programmed to form a system, and the feature extraction method of the roof image of the empty roof area can be processed automatically without manual intervention in the whole process. How to support the intervention to provide a better data basis, and through a variety of algorithm processing, to achieve effective calculation and processing of the roof image of the empty roof area, can effectively adjust and eliminate the brightness and noise of the picture, and extract the rock mass structure Surface profile, refine the profile of the structure surface and connect the interval points, obtain the skeleton of the structure surface, and then calculate the characteristic parameters of the structure surface, including the width and roughness of the crack gap, to classify the bare roof of the roadway.

In a specific embodiment, the adjustment of the position of the camera lens in S1 makes the camera lens and the ceiling remain relatively parallel, that is, the picture captured by the camera lens is flat, and when the camera lens is shooting, the camera is kept The lens is balanced to prevent shake.

In the above implementation process, the camera lens can take a flat picture, which effectively improves the subsequent processing process of the picture and maintains the accuracy of the picture.

In a specific embodiment, when the camera lens is shooting, the supplementary light is used to illuminate the empty ceiling, and the supplementary light is illuminated by shadowless LED lights.

In the above implementation process, clear pictures are taken in a relatively dark coal lane, and shadows are reduced to improve the accuracy of the picture processing process.

In a specific embodiment, when the computer in S2 collects the bare-top image, it transmits through a storage data card or a USB transmission method, and when the camera stores the bare-top image, it performs Compressed storage.

In the above implementation process, the coal lane is deep, the cable transmission is more wasteful of resources, and the wireless transmission signal is not good, which may easily cause the picture to lose data.

In a specific implementation, when the computer receives the picture through the storage data card or USB transmission, it first transmits the picture to the computer, then performs lossless decoding processing on the picture, and then preprocesses the picture.

In the above implementation process, after a picture is taken, the picture is compressed and stored, which can increase the storage capacity, and the decoding process is convenient to realize the processing operation of the picture.

In a specific implementation, the gamma correction in S3 is used to perform nonlinear operations or inverse operations on the luminance or tristimulus values of light in a film or imaging system, and is determined by the following Defined by the power law formula,

V _out =AV _in ^λ ,

where A is a constant, and the input and output values are non-negative real values. In the usual case of A=1, the input and output values range from 0 to 1, and the gamma value γ<1 When the gamma value is called encoding gamma value, the process of performing the above-mentioned power law for this encoding operation is also called gamma compression; on the contrary, when the gamma value γ>1 is called decoding gamma value, and performing this The above-mentioned power-law process used in the decoding operation is also called gamma expansion. The gamma correction has the following formulas for calculating lightness and grayscale:

RGB brightness calculation formula:

The value range of L is 0-1,

RGB grayscale calculation formula:

In the above implementation process, effective color processing of the picture can be achieved through gamma correction, the color and brightness of the image can be improved, and subsequent noise removal can be facilitated.

In a specific implementation, the median filter in S3 is to sort the gray values of pixels in a sliding window, and use the median value to replace the original gray value of the center pixel of the window, so as to realize the image Perform noise removal processing, and the steps of the median filter are as follows:

S301, roaming the filter template, namely a sliding window containing several points, in the image, and overlapping the template center with a certain pixel position in the figure;

S302, read the gray value of each corresponding pixel in the template;

S303, arrange these grayscale values from small to large;

S304. Take the middle data of this column of data, and assign it to the pixel at the center position of the corresponding template. If there are an odd number of elements in the window, the median value is the gray value of the middle element after the elements are sorted according to the size of the gray value. If there are an even number of elements, the median value is the average of the gray levels of the two middle elements after the elements are sorted by gray value.

In the above implementation process, the median filter has a good suppression effect on the impulse interference level salt and pepper noise, and can effectively protect the edges from blurring while suppressing random noise.

In a specific embodiment, the step of region growth segmentation in the S4 is as follows:

S401, scan the image sequentially, find the first pixel that has not yet been attributed, and set the pixel to be (x0, y0);

S402. Taking (x0, y0) as the center, consider the 8-neighborhood pixel (x, y) of (x0, y0), if (x0, y0) satisfies the growth criterion, combine (x, y) with (x0, y0) In the same area, push (x, y) onto the stack at the same time;

S403, take out a pixel from the stack and return it to S402 as (x0, y0);

S404, when the stack is empty, return to S401;

S405, repeating S401-S404 until each point in the image has an attribution, and the growth ends.

In the above implementation process, the region growing method is a method of gathering pixels according to the similar properties of pixels in the same object region, and merging adjacent pixels or other regions with the same properties into the current region to gradually grow area.

In a specific embodiment, the Hough transform in S4 is used to detect the boundary shape of the discontinuous point, and the fitting of the straight line and the curve is realized by transforming the image coordinate space into the parameter space;

Linear Coordinate Parameter Space

In the image x-y coordinate space, the line passing through the point (xi, yi) is represented as:

yi=axi+b,

Among them, the parameter a is the slope, b is the intercept,

There are an infinite number of straight lines passing through the point (xi, yi), which correspond to different a and b values,

If xi and yi are regarded as constants, and the original parameters a and b are regarded as variables, the formula is expressed as:

b=-xia+yi,

In this way, it is transformed to the parameter plane a-b, which is the Hough transform of the (xi, yi) point in Cartesian coordinates;

Polar coordinate parameter space

In polar coordinates, a straight line is represented by the following parametric equation,

ρ=xcosθ+ysinθ,

Among them, ρ represents the vertical distance from the line to the origin, θ represents the angle from the x-axis to the vertical line of the line, and the value range is ±90°;

Curve detection

The Hough transform is also suitable for curve detection where the equation is known,

For a known circle equation, the general equation for its rectangular coordinates is:

(x-a)2+(y-b)2=r2,

Among them, (a, b) are the coordinates of the center of the circle, and r is the radius of the circle;

Then, the parameter space can be expressed as (a, b, r), a circle in the image coordinate space corresponds to a point in the parameter space;

Detection of Arbitrary Shapes

First select any point (a, b) in the shape as the reference point, then from each point on the edge of the arbitrary shape, calculate its tangent direction φ and the offset vector r to the position of the reference point (a, b), and The angle α between r and the x-axis;

The position of the reference point (a, b) can be calculated by the following formula:

a=x+r(φ)cos(α(φ))

b=x+r(φ)sin(α(φ)).

In the above implementation process, the Hough transform realizes the detection of the discontinuous point boundary shape of the picture, and the fitting of the straight line and the curve is realized by transforming the image coordinate space into the parameter space.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present application more clearly, the following briefly introduces the accompanying drawings used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore do not It should be regarded as a limitation of the scope, and for those of ordinary skill in the art, other related drawings can also be obtained according to these drawings without any creative effort.

1 is a schematic flowchart of method steps provided by an embodiment of the present application;

2 is a schematic flowchart of steps of median filtering provided by an embodiment of the present application;

FIG. 3 is a schematic flowchart of the steps of region growth and segmentation provided by an embodiment of the present application.

Detailed ways

The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.

In order to make the purposes, technical solutions and advantages of the embodiments of the present application more clear, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments This is a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

Thus, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present application.

It should be noted that like numerals and letters refer to like items in the following figures, so once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

In the description of this application, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "upper", "lower", "front", " rear, left, right, vertical, horizontal, top, bottom, inside, outside, clockwise, counterclockwise, etc., or The positional relationship is based on the orientation or positional relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, Therefore, it should not be construed as a limitation on this application.

In addition, the terms "first" and "second" are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature defined as "first" or "second" may expressly or implicitly include one or more of that feature. In the description of the present application, "plurality" means two or more, unless otherwise expressly and specifically defined.

In this application, unless otherwise expressly specified and limited, the terms "installed", "connected", "connected", "fixed" and other terms should be understood in a broad sense, for example, it may be a fixed connection or a detachable connection , or integrated; it can be directly connected or indirectly connected through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between the two elements. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific situations.

In this application, unless otherwise expressly specified and defined, a first feature "on" or "under" a second feature may include direct contact between the first and second features, or may include the first and second features Not directly but through additional features between them. Also, the first feature being "above", "over" and "above" the second feature includes the first feature being directly above and obliquely above the second feature, or simply means that the first feature is level higher than the second feature. The first feature is "below", "below" and "below" the second feature includes the first feature being directly below and diagonally below the second feature, or simply means that the first feature has a lower level than the second feature.

Please refer to FIGS. 1-3 , the present application provides a method for extracting image features of a roof in a hollow roof area of a coal roadway, including the following method steps:

S1. After the fully mechanized excavator finishes cutting the coal body, adjust the position of the camera lens to aim at the exposed roof of the empty roof area;

S2. Use the well-connected computer to collect bare-top images;

S3. Preprocess the bare-top image, and use gamma correction and median filtering to adjust and eliminate the brightness and noise of the image;

S4. The image is extracted by region growth segmentation and Hough transform to extract the contour of rock mass structure surface;

S5. Refine the outline of the structural surface and connect the space points to obtain the skeleton of the structural surface;

S6. Calculate the characteristic parameters of the structure surface, including the width and roughness of the crack gap, and classify the bare roof of the roadway.

In the above implementation process, the method can be programmed to form a system, and the feature extraction method of the roof image of the empty roof area can be processed automatically without manual intervention in the whole process. How to support the intervention to provide a better data basis, and through a variety of algorithm processing, to achieve effective calculation and processing of the roof image of the empty roof area, can effectively adjust and eliminate the brightness and noise of the picture, and extract the rock mass structure Surface profile, refine the profile of the structure surface and connect the interval points, obtain the skeleton of the structure surface, and then calculate the characteristic parameters of the structure surface, including the width and roughness of the crack gap, to classify the bare roof of the roadway.

In the embodiment, the adjustment of the position of the camera lens in S1 makes the camera lens remain relatively parallel to the ceiling, that is, the image captured by the camera lens is flat, and when the camera lens is shooting, the balance of the camera lens is maintained to prevent Shaking enables the camera lens to take a flat picture, effectively improving the subsequent processing of the picture and maintaining the accuracy of the picture.

In a specific scheme, when the camera lens is shooting, the supplementary light is used to achieve supplementary lighting for the empty ceiling, and the supplementary light is illuminated by shadowless LED lights, and the shooting is clear in a relatively dark coal lane. picture, and reduce shadows, improve the accuracy of the picture processing process.

In the embodiment, when the computer in S2 collects the bare-top image, it transmits through a storage data card or a USB transmission method, and when the camera stores the bare-top image, it compresses and stores the bare-top image, and the The lane is deep, the cable transmission is a waste of resources, and the wireless transmission signal is not good, which may easily cause the loss of data in the picture. The accuracy of the picture can be maintained through the storage data card or USB transmission.

In a specific implementation, when the computer receives a picture through a storage data card or a USB transmission, it first transmits the picture to the computer, then performs lossless decoding processing on the picture, and then preprocesses the picture, before taking the picture. Afterwards, pressurized storage of the picture can increase the storage capacity, and decoding processing facilitates the realization of the picture processing operation.

In a specific implementation, the gamma correction in S3 is used to perform nonlinear operations or inverse operations on the luminance or tristimulus values of light in a film or imaging system, and is determined by the following Defined by the power law formula,

V _out =AV _in ^λ ,

where A is a constant, and the input and output values are non-negative real values. In the usual case of A=1, the input and output values range from 0 to 1, and the gamma value γ<1 is called encoding the gamma value, and the process of performing the above-mentioned power law of the encoding operation is also called gamma compression; on the contrary, when the gamma value γ>1 is called decoding the gamma value, and performing this The above-mentioned power-law process used in the decoding operation is also called gamma expansion. The gamma correction has the following formulas for calculating lightness and grayscale:

RGB brightness calculation formula:

The value range of L is 0-1,

RGB grayscale calculation formula:

Through gamma correction, effective color processing can be achieved on the picture, which improves the color and brightness of the image, and facilitates subsequent noise removal.

In an embodiment, the median filter in S3 is to sort the gray values of pixels in a sliding window, and use the median value to replace the original gray value of the center pixel of the window, so as to realize the denoising process of the image. , the steps of the median filter are as follows:

S301, roaming the filter template, namely a sliding window containing several points, in the image, and overlapping the template center with a certain pixel position in the figure;

S302, read the gray value of each corresponding pixel in the template;

S303, arrange these grayscale values from small to large;

S304. Take the middle data of this column of data, and assign it to the pixel at the center position of the corresponding template. If there are an odd number of elements in the window, the median value is the gray value of the middle element after the elements are sorted according to the size of the gray value. If there are an even number of elements, the median value is the average of the gray levels of the two middle elements after the elements are sorted by gray value.

The median filter has a good suppression effect on the impulse interference level salt and pepper noise, and can effectively protect the edges from blurring while suppressing random noise.

In an advantageous solution, the step of region growing segmentation in S4 is as follows:

S401, scan the image sequentially, find the first pixel that has not been attributed, and set the pixel to be (x0, y0);

S402. Taking (x0, y0) as the center, consider the 8-neighborhood pixel (x, y) of (x0, y0), if (x0, y0) satisfies the growth criterion, combine (x, y) with (x0, y0) In the same area, push (x, y) onto the stack at the same time;

S403, take out a pixel from the stack and return it to S402 as (x0, y0);

S404, when the stack is empty, return to S401;

S405, repeating S401-S404 until each point in the image has an attribution, and the growth ends.

The region growing method is a method of gathering pixels according to the similar properties of pixels in the same object region, and merges adjacent pixels or other regions with the same properties into the current region to gradually increase the region.

Specifically, the Hough transform in the S4 is used to detect the boundary shape of the discontinuous point, and the fitting of the straight line and the curve is realized by transforming the image coordinate space into the parameter space;

Linear Coordinate Parameter Space

In the image x-y coordinate space, the line passing through the point (xi, yi) is represented as:

yi=axi+b,

Among them, the parameter a is the slope, b is the intercept,

There are an infinite number of straight lines passing through the point (xi, yi), which correspond to different a and b values,

If xi and yi are regarded as constants, and the original parameters a and b are regarded as variables, the formula is expressed as:

b=-xia+yi,

In this way, it is transformed to the parameter plane a-b, and this transformation is the Hough transformation of the (xi, yi) point in Cartesian coordinates;

Polar coordinate parameter space

In polar coordinates, a straight line is represented by the following parametric equation,

ρ=xcosθ+ysinθ,

Among them, ρ represents the vertical distance from the line to the origin, θ represents the angle from the x-axis to the vertical line of the line, and the value range is ±90°;

Curve detection

The Hough transform is also suitable for curve detection where the equation is known,

For a known circle equation, the general equation for its rectangular coordinates is:

(x-a)2+(y-b)2=r2,

Among them, (a, b) are the coordinates of the center of the circle, and r is the radius of the circle;

Then, the parameter space can be expressed as (a, b, r), a circle in the image coordinate space corresponds to a point in the parameter space;

Detection of Arbitrary Shapes

First select any point (a, b) in the shape as the reference point, then from each point on the edge of the arbitrary shape, calculate its tangent direction φ and the offset vector r to the position of the reference point (a, b), and The angle α between r and the x-axis;

The position of the reference point (a, b) can be calculated by the following formula:

a=x+r(φ)cos(α(φ))

b=x+r(φ)sin(α(φ)).

The Hough transform realizes the detection of the discontinuous point boundary shape of the picture, and realizes the fitting of the straight line and the curve by transforming the image coordinate space into the parameter space.

The specific implementation steps are as follows:

The first step, after the fully mechanized excavator has finished cutting the coal body, adjust the position of the camera lens to aim at the exposed roof of the empty roof area;

The second step is to use the well-connected computer to collect bare-top images;

The third step is to preprocess the bare-top image, and use gamma correction and median filtering to adjust and eliminate the brightness and noise of the image;

The fourth step, the image is extracted by region growth segmentation and Hough transform to extract the outline of rock mass structure surface;

The fifth step is to refine the outline of the structural surface and connect the interval points to obtain the skeleton of the structural surface;

The sixth step is to calculate the characteristic parameters of the structure surface, including the width and roughness of the crack gap, and classify the bare roof of the roadway.

The above descriptions are merely examples of the present application, and are not intended to limit the protection scope of the present application. For those skilled in the art, various modifications and changes may be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of this application shall be included within the protection scope of this application. It should be noted that like numerals and letters refer to like items in the following figures, so once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

The above are only specific embodiments of the present application, but the protection scope of the present application is not limited to this. should be covered within the scope of protection of this application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims (10)

1. A method for extracting image features of a top plate of a coal roadway empty-top area is characterized by comprising the following steps:

s1, after the fully mechanized coal cutter cuts the coal body, adjusting the position of a camera lens to align to the exposed top plate of the empty roof area;

s2, collecting a bare top image by using an aboveground connection computer;

s3, preprocessing the bare top image, and respectively adjusting and eliminating the brightness and noise of the image by gamma correction and median filtering;

s4, extracting the contour of the rock mass structural plane from the picture through region growing segmentation and Hough transformation;

s5, thinning the structural surface outline and connecting the interval points to obtain a structural surface framework;

s6, calculating structural surface characteristic parameters including crack gap width and roughness, and classifying the roadway bare top.

2. The method for extracting image features of the roof of the coal roadway goaf roof area according to claim 1, wherein the position of the camera lens in S1 is adjusted so that the camera lens is parallel to the goaf roof, i.e. the picture taken by the camera lens is flat, and the camera lens is kept balanced and prevented from shaking when the camera lens takes a picture.

3. The method for extracting image features of the roof in the coal roadway goaf-top area according to claim 2, wherein when the camera lens is used for shooting, supplementary lighting is performed on the goaf top through a supplementary lighting lamp, and the supplementary lighting is performed by using a shadowless LED lamp.

4. The method for extracting roof image features of a coal roadway goaf-top area according to claim 1, wherein the computer in S2 transmits the image via a memory data card or USB transmission method when acquiring the image, and the camera compresses and stores the image when storing the image.

5. The method for extracting image features of the roof of the coal roadway goaf-top area as claimed in claim 4, wherein when the computer receives the pictures through the data card or USB transmission, the pictures are transmitted to the computer, then the pictures are subjected to lossless decoding processing, and then the pictures are preprocessed.

6. The method as claimed in claim 1, wherein the gamma correction in S3 is used to perform a nonlinear operation or inverse operation on the brightness or tristimulus values of light in a film or image system, which is defined by the following power law formula,

where a is a constant, the input and output values are all non-negative real values, in the normal case where a is 1, the input and output values range from 0 to 1, and the case where gamma is <1 is called encoding gamma, and the process using the above power law to perform this encoding operation is also called gamma compression; in contrast, the case where the gamma value γ >1 is referred to as a decoded gamma value, and the process using the above power law for performing this decoding operation is also referred to as gamma expansion.

7. The method for extracting image features of the top plate of the coal roadway goaf-top area according to claim 6, wherein the gamma correction is as follows for brightness and gray scale calculation formula:

RGB lightness calculation formula:

the value of L ranges from 0 to 1,

RGB gray scale calculation formula:

8. the method for extracting roof image features in a coal roadway goaf roof area according to claim 1, wherein the median filtering in S3 is to sort the gray values of pixels in a sliding window, and replace the original gray value of the central pixel of the window with the value thereof for implementing denoising processing on the image, and the steps of the median filtering are as follows:

s301, roaming a filtering template, namely a sliding window containing a plurality of points in an image, and overlapping the center of the template with a certain pixel position in the image;

s302, reading the gray value of each corresponding pixel in the template;

s303, arranging the gray values from small to large;

s304, taking the middle data of the data column, assigning the middle data to the pixel corresponding to the center position of the template, if the window has odd elements, taking the gray value of the middle element after the elements are sorted according to the gray value, if the window has even elements, taking the average value of the gray values of the middle two elements after the elements are sorted according to the gray value.

9. The method for extracting image features of the roof of the coal roadway goaf-top area according to claim 1, wherein the step of region growing and segmenting in S4 is as follows:

s401, sequentially scanning the image, finding out the 1 st pixel which is not attributed yet, and setting the pixel as (x0, y 0);

s402, centering on (x0, y0), considering (x0, y0) 8 neighborhood pixels (x, y), if (x0, y0) satisfies the growing criterion, (x, y) and (x0, y0) are merged in the same region, and (x, y) is pushed into the stack at the same time;

s403, taking a pixel out of the stack, and returning the pixel to S402 as (x0, y 0);

s404, returning to S401 when the stack is empty;

s405, repeating S401-S404 until each point in the image has attribution, and ending the growth.

10. The method for extracting image features of the roof of the coal roadway goaf-top area according to claim 1, wherein the hough transform in S4 is used for detecting the shape of the boundary of the discontinuity, and the fitting of a straight line and a curve is realized by transforming an image coordinate space into a parameter space;

linear coordinate parameter space

In the image x-y coordinate space, the straight line passing through the points (xi, yi) is represented as:

yi＝axi+b，

wherein the parameter a is the slope, b is the intercept,

the straight line passing through the points (xi, yi) has innumerable bars, and corresponds to different values of a and b,

if xi and yi are regarded as constants and the original parameters a and b are regarded as variables, the equation is expressed as:

b＝-xia+yi，

thus, the transformation is carried out to the parameter plane a-b, and the transformation is the Hough transformation of the (xi, yi) point in the rectangular coordinate;

polar coordinate parameter space

A straight line is expressed in polar coordinates by the following parametric equation,

ρ＝xcosθ+ysinθ，

wherein rho represents the vertical distance from the straight line to the original point, theta represents the angle from the x axis to the perpendicular line of the straight line, and the value range is +/-90 degrees;

curve detection

The Hough transform is equally applicable to curve detection where the equations are known,

for a known circular equation, the general equation for its rectangular coordinates is:

(x-a)2+(y-b)2＝r2，

wherein, (a, b) is the coordinate of the centre of a circle, r is the radius of the circle;

then, the parameter space can be represented as (a, b, r), with one circle in the image coordinate space corresponding to one point in the parameter space;

detection of arbitrary shapes

Firstly, selecting any point (a, b) in the shape as a reference point, and then calculating the tangential direction phi, the offset vector r from the position of the reference point (a, b) and the included angle alpha between r and the x axis from each point of the edge of the graph with the any shape;

the position of the reference point (a, b) can be calculated by:

a＝x+r(φ)cos(α(φ))

b＝x+r(φ)sin(α(φ))。

Images