WO2023092965A1 - 矿石图像的分割方法、装置及计算机可读存储介质 - Google Patents
矿石图像的分割方法、装置及计算机可读存储介质 Download PDFInfo
- Publication number
- WO2023092965A1 WO2023092965A1 PCT/CN2022/094213 CN2022094213W WO2023092965A1 WO 2023092965 A1 WO2023092965 A1 WO 2023092965A1 CN 2022094213 W CN2022094213 W CN 2022094213W WO 2023092965 A1 WO2023092965 A1 WO 2023092965A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- pixel
- target
- point
- ore
- image
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 75
- 238000003709 image segmentation Methods 0.000 title claims abstract description 15
- 230000011218 segmentation Effects 0.000 claims abstract description 41
- 230000008569 process Effects 0.000 claims description 27
- 230000003628 erosive effect Effects 0.000 claims description 16
- 238000010586 diagram Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000004891 communication Methods 0.000 description 6
- 239000004575 stone Substances 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000006835 compression Effects 0.000 description 2
- 238000007906 compression Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005530 etching Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000005260 corrosion Methods 0.000 description 1
- 230000007797 corrosion Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 229910052500 inorganic mineral Inorganic materials 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 239000011707 mineral Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration using local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
Definitions
- the present application relates to the technical field of ore segmentation, in particular to a method, device, and computer-readable storage medium for ore image segmentation.
- the ore sorting machine In the ore sorting machine, the ore is transported through the conveyor belt, and then a camera or X-ray is used to take pictures of the ore to obtain a photo image of the physical information and position of the ore. Due to factors such as the position and shooting angle of the camera or X-ray source, imaging cannot be guaranteed The subsequent stones are all independent of each other, and there will be some overlap. After overlapping, it will bring certain difficulties to the accurate and efficient classification of ores. Therefore, it is necessary to design an algorithm to segment the obtained overlapping ore images, and strive to separate each stone on the image without overlapping. Usually, concave points are used. Matching, watershed algorithm, semantic segmentation and other methods to segment the image.
- the main purpose of the present application is to provide a method, device, and computer-readable storage medium for ore image segmentation, aiming at solving the problem of over-segmentation easily caused by ore image segmentation in the prior art.
- the application provides a method for segmenting an ore image, the steps of the method for segmenting an ore image include:
- the saddle point is the maximum value pixel point on the target direction and the target direction corresponding A collection of minimum pixel points in the vertical direction;
- a dividing line is determined according to the position information of the saddle point and the target direction, and the ore is divided by the dividing line.
- the step of obtaining the grayscale mountain map of the ore includes:
- the ore image is converted into the grayscale mountain map.
- the pixel points include overlapping area pixel points
- the step of determining the saddle point in the overlapping area and the target direction corresponding to the saddle point according to the pixel values of each pixel point in the grayscale mountain map includes:
- the saddle point and the target direction are determined according to the pixel points in the target overlapping area.
- the step of obtaining the pixel points in the overlapping area satisfying preset rules among the pixel points in the overlapping area according to the pixel value, the extension line and the vertical line includes:
- the target pixel value is greater than the first pixel value and less than the second pixel value, and the first pixel value and the second pixel value are symmetrical about the target pixel value
- the second pixel value increases stepwise, the center point corresponding to the target pixel value is used as the pixel point of the target overlapping area, and the extension line is determined as the target extension line.
- the step of determining the saddle point and the target direction according to the pixel points of the target overlapping area includes:
- the overlapping area includes at least one saddle point
- the step of determining the dividing line according to the position information of the saddle point and the target direction includes:
- the distance between the saddle points is less than or equal to a preset distance threshold, and the azimuth angle difference is less than or equal to a preset angle threshold, calculating the coordinate mean value of all the saddle points according to the position information of each of the saddle points and according to the Calculate the azimuth angle of the target direction to obtain the mean value of the azimuth angle;
- a dividing line is determined based on the dividing point and the dividing direction.
- the step of converting the ore image into the grayscale mountain map includes:
- Preprocessing the ore image to obtain a binarized image corresponding to the ore image includes at least one of binarization processing and/or denoising processing;
- the superimposed binarized image is determined as the grayscale mountain map.
- the present application also provides a device for segmenting ore images.
- the device for segmenting ore images includes: a memory, a processor, and an ore image stored in the memory and operable on the processor.
- An image segmentation program when the ore image segmentation program is executed by the processor, the steps of the above-mentioned ore image segmentation method are realized.
- the present application also provides a computer-readable storage medium, the computer-readable storage medium stores a segmentation program of an ore image, and when the segmentation program of an ore image is executed by a processor, the above-mentioned The steps of the segmentation method of the ore image described above.
- a method, device, and computer-readable storage medium for ore image segmentation proposed in the embodiments of the present application obtain the grayscale mountain map corresponding to the ore, and then determine the saddle point in the overlapping area and the saddle point corresponding to the saddle point according to the grayscale mountain map.
- the target direction and then segment the ore image corresponding to the ore according to the saddle point and the target direction, thereby realizing accurate segmentation of the ore and solving the problem that traditional segmentation algorithms easily lead to over-segmentation.
- Fig. 1 is a schematic diagram of the terminal structure of the hardware operating environment involved in the solution of the embodiment of the present application;
- Fig. 2 is the schematic flow chart of the first embodiment of the segmentation method of the ore image of the present application
- FIG. 3 is a schematic diagram of the refinement process of step S10 of the first embodiment of the ore image segmentation method of the present application;
- Fig. 4 is a schematic diagram of the refinement process of step S12 of the first embodiment of the ore image segmentation method of the present application;
- Fig. 5 is the example diagram of the gray scale mountain diagram of the first embodiment of the segmentation method of the ore image of the present application
- Fig. 6 is an example diagram of a segmented ore image in the first embodiment of the method for segmenting an ore image of the present application
- FIG. 7 is a schematic diagram of the refinement process of step S20 of the second embodiment of the ore image segmentation method of the present application.
- FIG. 8 is a schematic diagram of the refinement process of step S23 of the third embodiment of the ore image segmentation method of the present application.
- FIG. 9 is a schematic diagram of the refinement process of step S30 of the fourth embodiment of the ore image segmentation method of the present application.
- the main solution of the embodiment of the present application is: obtain the grayscale mountain map of the ore; determine the saddle point of the overlapping area and the target direction corresponding to the saddle point according to the pixel value of each pixel point in the grayscale mountain map, and the saddle point is A set of maximum value pixel points in the target direction and minimum value pixel points in the vertical direction corresponding to the target direction; determine a dividing line according to the position information of the saddle point and the target direction, and use the A dividing line divides the ore.
- FIG. 1 is a schematic diagram of a terminal structure of a hardware operating environment involved in the solution of the embodiment of the present application.
- the terminal of the embodiment of the present application can be a PC, and can also be a smart phone, a tablet computer, an e-book reader, an MP3 (Moving Picture Experts Group Audio Layer III, moving picture expert compression standard audio level 3) player, an MP4 (Moving Picture Experts Group Audio Layer IV, dynamic image expert compression standard audio layer 3) Players, portable computers and other mobile terminal devices with processing functions.
- MP3 Motion Picture Experts Group Audio Layer III, moving picture expert compression standard audio level 3
- MP4 Moving Picture Experts Group Audio Layer IV, dynamic image expert compression standard audio layer
- the terminal may include: a processor 1001 , such as a CPU, a network interface 1004 , a user interface 1003 , a memory 1005 , and a communication bus 1002 .
- the communication bus 1002 is set to realize connection and communication between these components.
- the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface.
- the network interface 1004 may include a standard wired interface and a wireless interface (such as a WI-FI interface).
- the memory 1005 can be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory.
- the memory 1005 may also be a storage device independent of the aforementioned processor 1001 .
- the terminal may further include a camera, an RF (Radio Frequency, radio frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like.
- sensors such as light sensors, motion sensors and other sensors.
- the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and/or backlight.
- the gravitational acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when it is stationary, and can be used for applications that recognize the posture of mobile terminals (such as horizontal and vertical screen switching, Related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tap), etc.; of course, the mobile terminal can also be equipped with other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. No longer.
- terminal structure shown in FIG. 1 does not constitute a limitation on the terminal, and may include more or less components than those shown in the figure, or combine some components, or arrange different components.
- the memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a method for segmenting ore images.
- the network interface 1004 is mainly configured to connect to the background server and perform data communication with the background server;
- the user interface 1003 is mainly configured to connect to the client (client) and perform data communication with the client;
- the processor 1001 can be set to call the segmentation method of the ore image stored in memory 1005, and perform the following operations:
- the saddle point is the maximum value pixel point on the target direction and the target direction corresponding A collection of minimum pixel points in the vertical direction;
- a dividing line is determined according to the position information of the saddle point and the target direction, and the ore is divided by the dividing line.
- processor 1001 can call the segmentation method of the ore image stored in the memory 1005, and also perform the following operations:
- the ore image is converted into the grayscale mountain map.
- processor 1001 can call the segmentation program of the ore image stored in the memory 1005, and also perform the following operations:
- the saddle point and the target direction are determined according to the pixel points in the target overlapping area.
- processor 1001 can call the segmentation program of the ore image stored in the memory 1005, and also perform the following operations:
- the target pixel value is greater than the first pixel value and less than the second pixel value, and the first pixel value and the second pixel value are symmetrical about the target pixel value
- the second pixel value increases stepwise, the center point corresponding to the target pixel value is used as the pixel point of the target overlapping area, and the extension line is determined as the target extension line.
- processor 1001 can call the segmentation program of the ore image stored in the memory 1005, and also perform the following operations:
- processor 1001 can call the segmentation program of the ore image stored in the memory 1005, and also perform the following operations:
- the distance between the saddle points is less than or equal to a preset distance threshold, and the azimuth angle difference is less than or equal to a preset angle threshold, calculating the coordinate mean value of all the saddle points according to the position information of each of the saddle points and according to the Calculate the azimuth angle of the target direction to obtain the mean value of the azimuth angle;
- a dividing line is determined based on the dividing point and the dividing direction.
- processor 1001 can call the segmentation program of the ore image stored in the memory 1005, and also perform the following operations:
- Preprocessing the ore image to obtain a binarized image corresponding to the ore image includes at least one of binarization processing and/or denoising processing;
- the superimposed binarized image is determined as the grayscale mountain map.
- the first embodiment of the present application provides a segmentation method of an ore image
- the segmentation method of the ore image includes:
- Step S10 obtaining the grayscale mountain map of the ore
- Step S20 Determine the saddle point of the overlapping area and the target direction corresponding to the saddle point according to the pixel values of each pixel point in the grayscale mountain image, the saddle point is the maximum value pixel point in the target direction and the target direction The set of minimum pixel points in the vertical direction corresponding to the direction;
- Step S30 determining a dividing line according to the position information of the saddle point and the target direction, and dividing the ore by the dividing line.
- industrial ore refers to ore that can be developed in industrial batches, and only industrial ore can generate economic benefits.
- Industrial ore grade that is, the amount of useful minerals in the ore, is an important indicator that affects the selection of mining methods.
- the ore sorting machine the ore is transported through the conveyor belt, and then a camera or X-ray is used to take pictures of the ore to obtain a photo image of the physical information and position of the ore.
- a segmentation method of an ore image is proposed to segment overlapping ores to obtain each segmented ore.
- the grayscale mountain image is converted from an ore image corresponding to the ore.
- the step S10 includes:
- Step S11 receiving the ore image collected by the camera
- Step S12 converting the ore image into the grayscale mountain image.
- the ore image is an image obtained by photographing the ore on the conveyor belt by an ore image acquisition device.
- the ore image acquisition device may be an industrial camera, or an X-ray, or a CCD line An array camera-based image acquisition device supplemented by lidar.
- the ore image is converted into a corresponding grayscale mountain image.
- the step S12 includes:
- Step S123 sequentially superimposing the pixel values of each pixel in the binarized image after each erosion process to form the superimposed binarized image
- Step S124 determining the superimposed binarized image as the grayscale mountain map.
- preprocessing is performed on the ore image, and the preprocessing manner includes at least one of denoising processing and/or binarization processing.
- the ore image is composed of pixels one by one. First, denoise the ore image to obtain a denoised ore image, and then perform denoising on the denoised ore image. Binarization processing to obtain a binarized image, that is, the gray value of the pixel on the binarized image is only 0 or 1, that is, the entire binarized image presents an obvious black and white effect, and the image composed of black and white pixels is obtained. Binarize the image.
- the method of the etching process is to use structural elements of a size of 3x3 to etch the binary image. It can be understood that in practical applications, structural elements of a size of 5x5 can also be used to perform the etching process. , to reduce the number of times of corrosion, so that the cutting performance is improved, and users can use it according to the situation.
- the pixel values of each pixel in the binarized image after each erosion process are sequentially Perform superposition to form the superimposed binarized image, and then determine the superimposed binarized image as the grayscale mountain map.
- FIG. 5 is an example of a grayscale mountain map picture.
- the formula for the pixel value of each pixel in the binarized image after each erosion process is the following formula.
- dst(x, y) represents the pixel value whose image coordinates are (x, y) in the binarized image after the superimposition
- srci(x, y) represents the value of the binary image after the ith erosion process.
- the pixel value whose image coordinates are (x, y) in the valued image, n represents the total number of erosions.
- the pixel value of each pixel point in the grayscale mountain image is superimposed on the pixel value after each erosion process, and then according to the The pixel value determines the saddle point in the grayscale mountain map.
- the saddle point includes a plurality of saddle points, and the saddle point may be in the overlapping area of the ore image, and may also be in the non-overlapping area of the ore image , in the embodiment of the present application, in order to accurately segment the ore image, after obtaining the saddle points of the grayscale mountain image, the saddle points located in the overlapping area are selected from the saddle points, and then according to the saddle points located in the overlapping area The saddle point defines a split line that passes through the overlap region to split the overlapping ore from the overlap region to prevent over-segmentation.
- the overlapping area in the grayscale mountain map is obtained, and then the saddle point located in the overlapping area is determined according to the pixels located in the overlapping area, and then The division line is determined according to the saddle points located in the overlapping area. In this way, by reducing the number of pixels, the amount of calculation for screening saddle points is reduced, thereby improving the segmentation efficiency of the ore image.
- the saddle point is further extracted from the skeleton points after the skeleton points of the grayscale mountain map are extracted according to the skeleton algorithm .
- the overlapping area of the grayscale mountain map is obtained, and the skeleton points of the overlapping area can be extracted according to the skeleton algorithm, and then the The saddle point is selected from the skeleton points in the overlapping region.
- the saddle point is a maximum value pixel point in the target direction and a minimum value pixel point in the vertical direction corresponding to the target direction
- the saddle point is a maximum value pixel point in the target direction point and is the minimum pixel point in the vertical direction corresponding to the target direction, that is, the pixel values of other pixel points in the target direction are lower than the pixel values corresponding to the saddle point, and the other pixel points in the vertical direction
- the pixel values of the pixel points are all higher than the pixel values corresponding to the saddle points.
- the saddle point when the overlapping region includes multiple saddle points, there may be multiple maximum pixel points in the target direction, and there may be multiple minimum value pixel points in the vertical direction corresponding to the target direction, At this time, the saddle point includes a plurality of maximum value pixel points in the target direction, or the saddle point includes a plurality of minimum value pixel points in the vertical direction in the target direction, wherein, in the target direction
- the pixel values of the plurality of maximum value pixels can be the same or different, and the pixel values of the minimum value pixels in the vertical direction can be the same or different.
- the division point of the division line is determined according to the position information of the saddle point, that is, the position of the saddle point is determined as the position of the division point, and The target direction is determined as the splitting direction of the splitting line, and then a splitting line is formed according to the splitting point and the splitting direction, and then the overlapping area of the ore image is split based on the splitting line, so as to realize the overlapping The ore is segmented.
- the division line is confirmed by obtaining the division point and the direction of the division line, and the overlapping area in the ore image is segmented according to all the division lines confirmed, so that the overlapping area of the ore image can be separated.
- the ore image of the separated single ore is obtained, as shown in FIG. 6 , and FIG. 6 shows an example diagram of the divided ore image.
- the ore image is denoised and/or binarized to obtain the grayscale mountain map corresponding to the ore image, and then according to the grayscale mountain map
- the pixel value of each pixel point in the image extracts the saddle point in the overlapping area, and determines the target direction corresponding to the saddle point, and then determines the position of the saddle point as the position of the segmentation point, and determines the target direction as the segmentation point
- the segmentation direction of the line and then segment the overlapping area according to the segmentation line to obtain a complete image of the segmented ore.
- the dividing line of the overlapping area can be determined, so that the overlapping area can be segmented according to the dividing line, and a complete image of a single ore can be obtained, which solves the traditional segmentation algorithm due to the ore image outline and grayscale. Irregularity leads to the problem of image over-segmentation, which improves the accuracy of ore image segmentation.
- the S20 includes:
- Step S21 obtaining overlapping area pixel points according to the pixel values of each pixel point in the grayscale mountain image
- Step S22 taking any overlapping area pixel point as the center point and the preset position as the starting position, constructing a plurality of extension lines with preset angles along the preset direction and vertical lines corresponding to the extension lines;
- Step S23 according to the pixel value, the extension line and the vertical line, obtain the target overlapping area pixel point in the overlapping area pixel point and the target extension line in the extension line;
- Step S24 determining the saddle point and the target direction according to the pixel points in the target overlapping area.
- the pixels in the overlapping area are pixels located in the overlapping area in the ore image, and the overlapping area includes a plurality of overlapping area pixels.
- the saddle point needs to acquire the target overlapping area pixel among the multiple overlapping area pixels, and then determine the target overlapping area pixel as the saddle point.
- the way to acquire the pixel points in the target overlapping area among the plurality of overlapping area pixel points is to take each of the overlapping area pixel points as the center point and the preset position as the starting position, and move along the preset direction Construct multiple extension lines with preset angles and the vertical lines corresponding to the extension lines.
- the preset position can be the x-axis or the y-axis, and the preset position can be
- the set position can be set by the user.
- the preset direction can be clockwise or counterclockwise.
- the preset angle can be 30 degrees, 20 degrees, or 45 degrees.
- the preset angle can be set by the user.
- the present application takes the preset position as the x-axis as the initial position, counterclockwise as the initial direction, and 30 degrees as the preset angle for example analysis.
- the target overlapping area in the pixel point in the overlapping area is obtained according to the pixel value, the extension line, and the vertical line pixel and the target extension line in the extension line.
- the S23 includes:
- the target pixel value is greater than the first pixel value and less than the second pixel value, and the first pixel value and the second pixel value are symmetrical about the target pixel value
- the second pixel value increases stepwise, the center point corresponding to the target pixel value is used as the pixel point of the target overlapping area, and the extension line is determined as the target extension line.
- the first other pixel points are pixels other than the central point among the pixel points passed by the extension line, and the second other pixel points are among the pixel points passed by the vertical line
- the central point corresponds to multiple extension lines and vertical lines, and the first other pixel points corresponding to each extension line are different, and the second other pixel points corresponding to each vertical line are different. Pixels are also different.
- the pixel value of each pixel point in the gray scale image is used to obtain the The first pixel value corresponding to the first other pixel point, the second pixel value corresponding to the second other pixel point, and the target pixel value of the center point, it can be understood that the first pixel values corresponding to different extension lines are different , the second pixel values corresponding to different vertical lines are also different.
- the center point corresponding to the target pixel value is used as the pixel point of the target overlapping area. It can be understood that the center point satisfies the following conditions:
- the Pi and Qi are the target pixel values of the central point
- the P1, P2, ... Pn-1, Pn are the first pixel values of the first other pixel points
- the Q1, Q2 , ... Qn-1, Qn is the second pixel value of the second other pixel point
- c is a constant.
- the center point is used as the pixel point of the target overlapping area, and the extension line is determined as the target extension line.
- the target extension line takes the pixel point of the target overlapping area as the center point, and the pixel values of other pixel points of the target extension line except the pixel point of the target overlapping area are lower than the pixel values of the pixel points of the target overlapping area, and the The pixel values of the other pixel points are symmetrical to the pixel values of the pixel points in the target overlapping area.
- the saddle point and the target direction are determined according to the pixel points in the target overlapping area.
- the step S24 includes:
- Step S241 obtaining the target angle between the target extension line and the starting position, and determining the target angle as the azimuth of the target extension line;
- Step S242 determining the pixel point in the target overlapping area as the saddle point, and determining the target direction according to the azimuth angle.
- the target extension line is the target extension line among the extension lines corresponding to the pixel points in the target overlapping area
- the pixel points in the target overlapping area are the maximum value pixel points on the target extension line and
- the minimum value pixel point on the target vertical line corresponding to the target extension line, the pixel value corresponding to other pixel points on the target extension line except the target overlapping area pixel point is higher than that of the target overlapping area pixel point
- the pixel values are low and symmetrical with respect to the pixel values of the pixel points in the target overlapping area.
- the included angle between the target extension line and the x-axis is acquired, and the included angle is determined as the azimuth of the target extension line, and then the The azimuth angle is determined as an azimuth angle corresponding to the target direction, so as to determine the target direction.
- the position of the pixel point in the target overlapping area is acquired, and the position of the pixel point in the target overlapping area is determined as the position of the saddle point.
- a plurality of extension lines with preset angles included along the preset direction and the corresponding vertical lines of the extension lines are constructed. line, and then according to the pixel value of each pixel in the grayscale mountain map, the extension line and the vertical line, the target overlapping area pixel in the overlapping area pixels is obtained, and then according to the target overlapping area pixel Determine the target extension line in the point, and then determine the position of the pixel point in the target overlapping area as the position of the saddle point, and determine the azimuth angle corresponding to the target extension line as the azimuth angle corresponding to the target direction, so as to determine the position of the saddle point.
- the target direction by constructing a plurality of extension lines centered on the pixel point in the overlapping area, and then judging whether the pixel point in the overlapping area is a pixel point in the target overlapping area, and then determining the saddle point and the target direction, and then using the saddle point And the target direction determines the dividing line, so as to divide the overlapping area according to the dividing line, so that the overlapping area can be divided according to the dividing line to obtain a complete image of a single ore, which solves the problem of the traditional segmentation algorithm due to the ore image outline and gray
- the irregularity of the degree leads to the problem of over-segmentation of the image, which improves the accuracy of the ore image segmentation.
- the overlapping region includes at least one saddle point
- the step S30 includes:
- Step S31 acquiring the distance between each of the saddle points and the azimuth difference between the azimuths corresponding to the target direction corresponding to each of the saddle points according to the position information of each of the saddle points;
- Step S32 determining that the distance between the saddle points is less than or equal to the preset distance threshold, and the azimuth angle difference is less than or equal to the preset angle threshold, and calculating the coordinate mean value of all the saddle points according to the position information of each of the saddle points And calculating the mean value of the azimuth angle according to the azimuth angle of the target direction;
- Step S33 determining the location of the mean value of the coordinates as the dividing point corresponding to the dividing line, and determining the dividing direction of the dividing line according to the mean value of the azimuth angle;
- Step S34 determining a dividing line based on the dividing point and the dividing direction.
- the application proposes a A way to determine the dividing line.
- the distance between each saddle point and the azimuth angle difference corresponding to the target direction corresponding to the saddle point are obtained according to the position information of the saddle point, and the position information is the The abscissa value and ordinate value corresponding to the saddle point, for example, there are saddle point A, saddle point B, saddle point C in the overlapping area A, the azimuth of the target direction corresponding to the saddle point A is G1, and the azimuth corresponding to the target direction corresponding to the saddle point B G2, the azimuth corresponding to the target direction corresponding to the saddle point C is G3, and obtain the distance L1 between the saddle point A and the saddle point B, the distance L2 between the saddle point A and the saddle point C, and the distance between the saddle point B and the saddle point C L3, and the azimuth difference is: G1-G2, G2-G3, G1-G3.
- the distance between each of the saddle points and the azimuth angle difference corresponding to the target direction corresponding to each of the saddle points it is judged whether the distance is less than or equal to a preset distance preset and the azimuth is judged Whether the angle difference is less than or equal to a preset angle threshold, the distance between the saddle points is less than or equal to a preset distance threshold, and the azimuth difference is less than or equal to a preset angle threshold, according to the position information of the saddle points Obtain the mean value of the coordinates corresponding to all the saddle points and obtain the mean value of the azimuth corresponding to the saddle point according to the target direction.
- the mean value of the coordinates includes the mean value of the abscissa and the mean value of the ordinate.
- saddle point C the position information of saddle point A is (x1, y1) and the azimuth angle of the corresponding target direction is G1
- the position information of saddle point B is (x2, y2) and the azimuth angle corresponding to the corresponding target direction is G2
- saddle point The location information of C is (x2, y3) and the azimuth angle corresponding to the corresponding target direction is G3, that is, the mean value of the abscissa is (x1+x2+x2)/3
- the mean value of the ordinate is (y1+y2+ y2)/3
- the mean value of the abscissa and the mean value of the ordinate determine that the coordinate mean value is ((x1+x2+x2)/3, (y1+y2+y2)/3), and the mean value of the azimuth angle is (G1+G2+G3)/3.
- the grayscale mountain map may include multiple overlapping areas, and each overlapping area may include multiple saddle points.
- each saddle point set includes at least two saddle point sets, the way of dividing the saddle point into multiple saddle point sets is to obtain the distance and azimuth angle difference between each saddle point,
- the distance between the saddle points is less than or equal to the preset distance threshold and the azimuth difference is less than or equal to the preset angle threshold, the distance is less than or equal to the preset distance threshold and the azimuth difference is less than or equal to the preset
- the angle threshold is determined as a saddle point in the same saddle point set.
- the distance between two saddle points in the same saddle point set is less than or equal to the preset distance threshold, and the saddle points in the same saddle point set correspond to The azimuth angle differences between the azimuth angles corresponding to the target directions are all less than or equal to the preset angle threshold.
- the coordinate mean value and the azimuth angle mean value corresponding to the saddle point set are determined according to the saddle points in each saddle point set, and the coordinate mean values corresponding to different saddle point sets are different and different
- the mean values of azimuth angles corresponding to the set of saddle points can be the same or different.
- the position of the coordinate mean value is determined as the division point corresponding to the division line, and the division line is determined according to the azimuth mean value
- the dividing direction is determined, and then the dividing line is determined based on the dividing point and the dividing direction, and then the overlapping area is divided according to the dividing line.
- the distance between each saddle point and the azimuth angle difference corresponding to the target direction corresponding to each saddle point are obtained, and when the distance is less than or equal to the preset distance threshold, and the azimuth angle difference is less than or equal to the preset angle threshold, obtain at least one saddle point whose distance is less than or equal to the preset distance threshold and the azimuth angle difference is less than or equal to the preset angle threshold, and then obtain the at least one The mean value of the coordinates and the mean value of the azimuth angle corresponding to the saddle point, and then determine the dividing line according to the mean value of the coordinate and the mean value of the azimuth angle, so as to avoid the problem of generating multiple dividing lines when the distance between the saddle points is too short, thereby reducing the amount of calculation , thus improving the segmentation efficiency of the ore image.
- the embodiment of the present application also proposes a computer-readable storage medium, the computer-readable storage medium stores an ore image segmentation program, and when the ore image segmentation program is executed by a processor, each of the above-mentioned Example steps.
- the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation.
- the technical solution of the present application can be embodied in the form of a software product in essence or the part that contributes to the prior art, and the computer software product is stored in a storage medium (such as ROM/RAM) as described above. , magnetic disk, optical disk), including several instructions to make a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) execute the methods described in various embodiments of the present application.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
本申请公开了一种矿石图像的分割方法、装置及计算机可读存储介质，应用于矿石分割技术领域，该方法包括：获取矿石的灰度山图；根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
Description
相关申请
本申请要求于2021年11月23号申请的、申请号为202111399105.5的中国专利申请的优先权，其全部内容通过引用结合于此。
本申请涉及矿石分割技术领域，尤其涉及矿石图像的分割方法、装置及计算机可读存储介质。
在矿石分选机中，矿石通过传送带进行传输，然后用相机或者X射线拍摄矿石照片，得到矿石物理信息和位置的照片图像，由于相机或者X射线源的位置和拍摄角度等因素，不能保证成像之后的石头全是相互独立开来，会存在一定的重叠。重叠之后会对于矿石的准确高效的分类带来一定的困难，因此需要设计一种算法对得到的重叠矿石图像进行分割，力求图像上每颗石头都是相互隔开不存在重叠，通常采用凹点匹配、分水岭算法、语义分割等方法对图像进行分割。但是对于矿石图像去重叠的分割，由于矿石图像轮廓和灰度的不规则性，以上方法难以找到准确的分割线，容易导致图像过分割，影响矿石图像分割效果。
上述内容仅用于辅助理解本申请的技术方案，并不代表承认上述内容是现有技术。
申请内容
本申请的主要目的在于提供一种矿石图像的分割方法、装置及计算机可读存储介质，旨在解决现有技术中分割矿石图像容易导致过分割的问题。
为实现上述目的，本申请提供一种矿石图像的分割方法，所述矿石图像的分割方法的步骤包括：
获取矿石的灰度山图；
根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；
根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
在一实施方式中，所述获取矿石的灰度山图的步骤包括：
接收摄像头采集的矿石图像；
将所述矿石图像转换成所述灰度山图。
在一实施方式中，所述像素点中包括重叠区域像素点，所述根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向的步骤包括：
根据所述灰度山图中的各个像素点的像素值获取重叠区域像素点；
以任一重叠区域像素点为中心点并以预设位置作起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线；
根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线；
根据所述目标重叠区域像素点确定所述鞍点以及所述目标方向。
在一实施方式中，所述根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中满足预设规则的目标重叠区域像素点的步骤包括：
根据所述像素值获取所述延伸线上除所述中心点以外的第一其他像素点的第一像素值、所述垂线上除所述中心点以外的第二其他像素点的第二像素值，以及所述中心点的目标像素值；
在所述目标像素值大于所述第一像素值并小于所述第二像素值第一像素值，且所述第一像素值和所述第二像素值关于所述目标像素值对称，所述第二像素值呈阶梯状递增时，将所述目标像素值对应的所述中心点作为所述目标重叠区域像素点，并将所述延伸线确定为目标延伸线。
在一实施方式中，所述根据所述目标重叠区域像素点确定鞍点以及所述目标方向的步骤包括：
获取所述目标延伸线与所述起始位置之间的目标夹角，将所述目标夹角确定为所述目标延伸线的方位角；
将所述目标重叠区域像素点确定为所述鞍点，并根据所述方位角确定所述目标方向。
在一实施方式中，所述重叠区域包括至少一个鞍点，所述根据所述鞍点的位置信息以及所述目标方向确定分割线的步骤包括：
根据各个所述鞍点的位置信息获取各个所述鞍点之间的距离以及各个所 述鞍点对应的目标方向对应的方位角之间的方位角差值；
确定所述鞍点之间的距离小于或等于预设距离阈值，且所述方位角差值小于或等于预设角度阈值，根据各个所述鞍点的位置信息计算所有所述鞍点的坐标均值以及根据所述目标方向的方位角计算得出方位角均值；
将所述坐标均值所在位置确定为所述分割线对应的分割点，根据所述方位角均值确定所述分割线的分割方向；
基于所述分割点以及所述分割方向确定分割线。
在一实施方式中，所述将所述矿石图像转换成所述灰度山图的步骤包括：
对所述矿石图像进行预处理，以获取所述矿石图像对应的二值化图像，所述预处理的方式包括二值化处理和/或去噪声处理的至少一种；
将所述二值化图像进行多次腐蚀处理，获取每次腐蚀处理后的所述二值化图像中各个像素点的像素值；
依次将各个所述像素值进行叠加，以形成叠加后的所述二值化图像；
将所述叠加后的所述二值化图像确定为所述灰度山图。
此外，为实现上述目的，本申请还提供一种矿石图像的分割装置，所述矿石图像的分割装置包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的矿石图像的分割程序，所述矿石图像的分割程序被所述处理器执行时实现如上所述的矿石图像的分割方法的步骤。
此外，为实现上述目的，本申请还提供一种计算机可读存储介质，所述计算机可读存储介质上存储有矿石图像的分割程序，所述矿石图像的分割程序被处理器执行时实现如上所述的矿石图像的分割方法的步骤。
本申请实施例提出的一种矿石图像的分割方法、装置及计算机可读存储介质，通过获取矿石对应的灰度山图，进而根据所述灰度山图确定重叠区域中的鞍点以及鞍点对应的目标方向，进而根据所述鞍点以及所述目标方向对所述矿石对应的矿石图像进行分割，从而实现了矿石的准确分割，解决了传统的分割算法容易导致过分割的问题。
图1是本申请实施例方案涉及的硬件运行环境的终端结构示意图；
图2为本申请矿石图像的分割方法第一实施例的流程示意图；
图3为本申请矿石图像的分割方法第一实施例步骤S10的细化流程示意图；
图4为本申请矿石图像的分割方法第一实施例步骤S12的细化流程示意图；
图5为本申请矿石图像的分割方法第一实施例的灰度山图示例图；
图6为本申请矿石图像的分割方法第一实施例中分割后的矿石图像的示例图；
图7为本申请矿石图像的分割方法第二实施例步骤S20的细化流程示意图；
图8为本申请矿石图像的分割方法第三实施例步骤S23的细化流程示意图；
图9为本申请矿石图像的分割方法第四实施例步骤S30的细化流程示意图。
本申请目的的实现、功能特点及优点将结合实施例，参照附图做进一步说明。
应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
本申请实施例的主要解决方案是：获取矿石的灰度山图；根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
如图1所示，图1是本申请实施例方案涉及的硬件运行环境的终端结构示意图。
本申请实施例终端可以是PC，也可以是智能手机、平板电脑、电子书阅读器、MP3(Moving Picture Experts Group Audio Layer III，动态影像专家压缩标准音频层面3)播放器、MP4(Moving Picture Experts Group Audio Layer IV，动态影像专家压缩标准音频层面3)播放器、便携计算机等具有处理功能的可移动式终端设备。
如图1所示，该终端可以包括：处理器1001，例如CPU，网络接口1004，用户接口1003，存储器1005，通信总线1002。其中，通信总线1002设置为实现这些组件之间的连接通信。用户接口1003可以包括显示屏(Display)、输入 单元比如键盘(Keyboard)，可选用户接口1003还可以包括标准的有线接口、无线接口。网络接口1004可选的可以包括标准的有线接口、无线接口(如WI-FI接口)。存储器1005可以是高速RAM存储器，也可以是稳定的存储器(non-volatile memory)，例如磁盘存储器。存储器1005可选的还可以是独立于前述处理器1001的存储装置。
在一实施方式中，终端还可以包括摄像头、RF(Radio Frequency，射频)电路、传感器、音频电路、WiFi模块等等。其中，传感器比如光传感器、运动传感器以及其他传感器。具体地，光传感器可包括环境光传感器及接近传感器，其中，环境光传感器可根据环境光线的明暗来调节显示屏的亮度，接近传感器可在移动终端移动到耳边时，关闭显示屏和/或背光。作为运动传感器的一种，重力加速度传感器可检测各个方向上(一般为三轴)加速度的大小，静止时可检测出重力的大小及方向，可用于识别移动终端姿态的应用(比如横竖屏切换、相关游戏、磁力计姿态校准)、振动识别相关功能(比如计步器、敲击)等；当然，移动终端还可配置陀螺仪、气压计、湿度计、温度计、红外线传感器等其他传感器，在此不再赘述。
本领域技术人员可以理解，图1中示出的终端结构并不构成对终端的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置。
如图1所示，作为一种计算机存储介质的存储器1005中可以包括操作系统、网络通信模块、用户接口模块以及矿石图像的分割方法。
在图1所示的终端中，网络接口1004主要设置为连接后台服务器，与后台服务器进行数据通信；用户接口1003主要设置为连接客户端(用户端)，与客户端进行数据通信；而处理器1001可以设置为调用存储器1005中存储的矿石图像的分割方法，并执行以下操作：
获取矿石的灰度山图；
根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；
根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割方法，还执行以下操作：
接收摄像头采集的矿石图像；
将所述矿石图像转换成所述灰度山图。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割程序，还执行以下操作：
根据所述灰度山图中的各个像素点的像素值获取重叠区域像素点；
以任一重叠区域像素点为中心点并以预设位置作起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线；
根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线；
根据所述目标重叠区域像素点确定所述鞍点以及所述目标方向。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割程序，还执行以下操作：
根据所述像素值获取所述延伸线上除所述中心点以外的第一其他像素点的第一像素值、所述垂线上除所述中心点以外的第二其他像素点的第二像素值，以及所述中心点的目标像素值；
在所述目标像素值大于所述第一像素值并小于所述第二像素值第一像素值，且所述第一像素值和所述第二像素值关于所述目标像素值对称，所述第二像素值呈阶梯状递增时，将所述目标像素值对应的所述中心点作为所述目标重叠区域像素点，并将所述延伸线确定为目标延伸线。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割程序，还执行以下操作：
获取所述目标延伸线与所述起始位置之间的目标夹角，将所述目标夹角确定为所述目标延伸线的方位角；
将所述目标重叠区域像素点确定为所述鞍点，并根据所述方位角确定所述目标方向。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割程序，还执行以下操作：
根据各个所述鞍点的位置信息获取各个所述鞍点之间的距离以及各个所述鞍点对应的目标方向对应的方位角之间的方位角差值；
确定所述鞍点之间的距离小于或等于预设距离阈值，且所述方位角差值小于或等于预设角度阈值，根据各个所述鞍点的位置信息计算所有所述鞍点的坐标均值以及根据所述目标方向的方位角计算得出方位角均值；
将所述坐标均值所在位置确定为所述分割线对应的分割点，根据所述方位角均值确定所述分割线的分割方向；
基于所述分割点以及所述分割方向确定分割线。
进一步地，处理器1001可以调用存储器1005中存储的矿石图像的分割程序，还执行以下操作：
对所述矿石图像进行预处理，以获取所述矿石图像对应的二值化图像，所述预处理的方式包括二值化处理和/或去噪声处理的至少一种；
将所述二值化图像进行多次腐蚀处理，获取每次腐蚀处理后的所述二值化图像中各个像素点的像素值；
依次将各个所述像素值进行叠加，以形成叠加后的所述二值化图像；
将所述叠加后的所述二值化图像确定为所述灰度山图。
参照图2，本申请第一实施例提供一种矿石图像的分割方法，所述矿石图像的分割方法包括：
步骤S10，获取矿石的灰度山图；
步骤S20，根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；
步骤S30，根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
在本实施例中，工业矿石是指可以工业化批量开发的矿石，只有工业矿石才能产生经济效益。工业矿石品位，即矿石中有用矿物含量的多少，它是影响采矿方法选择的一个重要指标。现有采矿业中，首先将矿山开采的原始较大矿石经过碎石机粉碎和除尘之后变成最大直径在2-8cm不规则小矿石，然后用矿石分选机进行分选。在矿石分选机中，矿石通过传送带进行传输，然后用相机或者X射线拍摄矿石照片，得到矿石物理信息和位置的照片图像，由于相机或者X射线源的位置和拍摄角度等因素，不能保证成像之后的石头全是相互独立开来，可能会存在一定的重叠，即矿石与矿石之间发生重叠，重叠之后会对于矿石的准确高效的分类带来一定的困难，基于此，本申请实施例提出了一种矿石图像的分割方法，以对发生重叠的矿石进行分割，以获取分割后的各个矿石。
在一实施方式中，所述灰度山图为所述矿石对应的矿石图像转换而成的。在一实施方式中，参照图3，所述步骤S10包括：
步骤S11，接收摄像头采集的矿石图像；
步骤S12，将所述矿石图像转换成所述灰度山图。
在一实施方式中，所述矿石图像为矿石图像采集装置对传送带上的矿石进行拍摄所得到的图像，所述矿石图像采集装置可以是工业相机，还可以是X射线，还可以是以CCD线阵相机为主，激光雷达为辅的图像采集装置。
在一实施方式中，在获取所述矿石图像后，将所述矿石图像转换成对应的灰度山图。
在一实施方式中，参照图4，所述步骤S12包括:
S121，对所述矿石图像进行预处理，以获取所述矿石图像对应的二值化图像，所述预处理的方式包括二值化处理和/或去噪声处理的至少一种；
S122，将所述二值化图像进行多次腐蚀处理，获取每次腐蚀处理后的所述二值化图像中各个像素点的像素值；
步骤S123，依次将每次腐蚀处理后的所述二值化图像中各个像素点的像素值进行叠加，以形成叠加后的所述二值化图像；
步骤S124，将所述叠加后的所述二值化图像确定为所述灰度山图。
在一实施方式中，在获取所述矿石图像后，对所述矿石图像进行预处理，所述预处理的方式包括去噪声处理和/或二值化处理的至少一种。在一实施方式中，所述矿石图像是由一个一个的像素点组成的，先对所述矿石图像进行去噪声处理，得到去噪后的矿石图像，进而对所述去噪后的矿石图像进行二值化处理，得到二值化图像，即二值化图像上的像素点的灰度值只为0或1，也就是将整个二值化图像呈现出明显的黑白效果，得到黑白像素组成的二值化图像。
在一实施方式中，在获取所述二值化图像后，对所述二值化图像进行多次腐蚀处理，以腐蚀所述二值化图像对应的边界，进而缩小所述二值化图像，具体地，所述腐蚀处理的方式为采用3x3大小的结构元素对所述二值化图像进行腐蚀，可以理解的是，在实际应用时也可以根据情况采用5x5大小的结构元素进行所述腐蚀处理，用以减少腐蚀次数，使得切割性能被提高，用户可以根据情况自行使用。
在一实施方式中，在获取每次腐蚀处理后的所述二值化图像中各个像素点的像素值后，依次将每次腐蚀处理后的所述二值化图像中各个像素点的像素值进行叠加，以形成叠加后的所述二值化图像，进而将所述叠加后的所述二值化图像确定为所述灰度山图，参照图5，图5为灰度山图的示例图。
在一实施方式中，所述将每次腐蚀处理后的所述二值化图像中各个像素点的像素值的公式为下列公式。
其中，dst(x，y)表示所述叠加后的所述二值化图像中图像坐标为(x，y)的像素值，srci(x，y)表示第i次腐蚀处理后的所述二值化图像中图像坐标为(x，y)的像素值，n表示腐蚀的总次数。
在一实施方式中，在获取所述灰度山图后，基于所述灰度山图中各个像素点的像素值为每次腐蚀处理后的像素值进行叠加后的像素值，进而根据所述像素值确定所述灰度山图中的鞍点，可以理解的是，所述鞍点包括多个鞍点，所述鞍点可能处于矿石图像的重叠区域内，还可能处于所述矿石图像的非重叠区域内，在本申请实施例中，为了准确分割所述矿石图像，在获取所述灰度山图的各个鞍点后，从所述各个鞍点筛选出位于重叠区域的鞍点，进而根据所述位于重叠区域的鞍点确定分割线，所述分割线经过重叠区域，以从重叠区域将发生重叠的矿石进行分割，以防止过分割。
在一实施方式中，为了减少计算量，在获取所述灰度山图后，获取所述灰度山图中的重叠区域，进而根据位于重叠区域内的像素点确定位于重叠区域的鞍点，进而根据所述位于重叠区域的鞍点确定分割线，如此，通过减少像素点的数量，进而减少筛选鞍点的计算量，从而提高矿石图像的分割效率。
在一实施方式中，在又一实施例中，在获取所述灰度山图后，根据骨架算法提取到所述灰度山图的骨架点后，进而从所述骨架点中提取所述鞍点。
在一实施方式中，在又一实施例中，在获取所述灰度山图后，获取所述灰度山图的重叠区域，可根据骨架算法提取到所述重叠区域的骨架点，进而在所述重叠区域的骨架点中筛选出所述鞍点。
在一实施方式中，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点，所述鞍点为目标方向上的极大值像素点且为所述目标方向对应的垂直方向上的极小值像素点，即所述目标方向上的其他像素点的像素值均比所述鞍点对应的像素值低，所述垂直方向上的其他像素点的像素值均比所述鞍点对应的像素值高。
在一实施方式中，在所述重叠区域包括多个鞍点时，所述目标方向上可以存在多个极大值像素点，所述目标方向对应的垂直方向上存在多个极小值像素点，此时，所述鞍点包括多个所述目标方向上的极大值像素点，或所述 鞍点包括多个所述目标方向上的垂直方向的极小值像素点，其中，所述目标方向上的多个极大值像素点的像素值可以相同，也可以不同，所述垂直方向上的极小值像素点的像素值可以相同，也可以不同。
在一实施方式中，在获取所述鞍点以及所述目标方向后，根据所述鞍点的位置信息确定所述分割线的分割点，即将所述鞍点所在位置确定为所述分割点所在位置，将所述目标方向确定为所述分割线的分割方向，进而根据所述分割点以及所述分割方向形成分割线，进而基于所述分割线分割所述矿石图像的重叠区域，以实现对重叠后的矿石进行分割，本申请实施例通过既获取分割点以及分割线的方向确认分割线，根据确认的所有的分割线对矿石图像中的重叠区域进行分割，可以实现矿石图像的重叠区域分离，就可以得到分离后的单个矿石的矿石图像，如图6所示，图6示出了分割后的矿石图像的示例图。
在本申请实施例中，在获取矿石图像后，对所述矿石图像进行去噪和/或二值化处理，以获取所述矿石图像对应的灰度山图，进而根据所述灰度山图中各个像素点的像素值提取重叠区域的鞍点，并确定所述鞍点对应的目标方向，进而将所述鞍点所在位置确定为所述分割点所在位置，并将所述目标方向确定为所述分割线的分割方向，进而根据分割线对重叠区域进行分割，得到分割后的矿石的完整图像。通过确定鞍点以及鞍点对应的目标方向进而确定重叠区域的分割线，这样便可以根据分割线对重叠区域进行分割，得到单个矿石的完整图像，解决了传统的分割算法因矿石图像轮廓和灰度的不规则性导致图像过分割的问题，提高了矿石图像分割的准确性。
在一实施方式中，基于第一实施例，参照图7，所述S20包括：
步骤S21，根据所述灰度山图中的各个像素点的像素值获取重叠区域像素点；
步骤S22，以任一重叠区域像素点为中心点并以预设位置作起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线；
步骤S23，根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线；
步骤S24，根据所述目标重叠区域像素点确定所述鞍点以及所述目标方向。
在本申请实施例中，所述重叠区域像素点为位于所述矿石图像中的重叠区域的像素点，所述重叠区域中包括多个重叠区域像素点，为了从多个重叠 区域像素点提取出鞍点，需获取所述多个重叠区域像素点中的目标重叠区域像素点，进而将所述目标重叠区域像素点确定为所述鞍点。
在一实施方式中，获取所述多个重叠区域像素点中的目标重叠区域像素点的方式为以各个所述重叠区域像素点为中心点并以预设位置做起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线，可以理解的是，一个重叠区域像素点对应的延伸线可以是多条，不同的重叠区域像素点对应的延伸线不同，所述垂线的中心点和所述延伸线的中心点相同，均为所述重叠区域像素点，其中，所述预设位置可以是x轴，还可以是y轴，所述预设位置可以是用户自行设定，所述预设方向可以是顺时针方向，也可以是逆时针方向，所述预设角度可以是30度，还可以20度，还可以是45度，所述预设角度可以是用户自行设定。
在一实施方式中，本申请以所述预设位置为x轴作为起始位置，以逆时针作起始方向，以30度作为预设角度进行举例分析。
在一实施方式中，在获取各个所述重叠区域像素点对应的延伸线以及垂线，根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线。
在一实施方式中，所述S23包括：
根据所述像素值获取所述延伸线上除所述中心点以外的第一其他像素点的第一像素值、所述垂线上除所述中心点以外的第二其他像素点的第二像素值，以及所述中心点的目标像素值；
在所述目标像素值大于所述第一像素值并小于所述第二像素值第一像素值，且所述第一像素值和所述第二像素值关于所述目标像素值对称，所述第二像素值呈阶梯状递增时，将所述目标像素值对应的所述中心点作为所述目标重叠区域像素点，并将所述延伸线确定为目标延伸线。
在一实施方式中，所述第一其他像素点为所述延伸线经过的像素点中除所述中心点以外的像素点，所述第二其他像素点为所述垂线经过的像素点中除所述中心点以外的像素点，可以理解的是，一个中心点对应着多条延伸线和垂线，每条延伸线对应的第一其他像素点不同，每条垂线对应的第二其他像素点也不同。
在一实施方式中，在确定各个所述延伸线上的第一其他像素点以及所述垂线上的第二其他像素点后，根据所述灰度图中各个像素点的像素值获取所述第一其他像素点对应的第一像素值、所述第二其他像素点对应的第二像素 值以及所述中心点的目标像素值，可以理解的是，不同延伸线对应的第一像素值不同，不同的垂线对应的第二像素值也不同。
在一实施方式中，在获取每条延伸线上各自对应的所述第一像素值和每条垂线上各自对应的第二像素值后，比对所述目标像素值以及所述第一像素值和所述第二像素值，以及判断所述第一像素值和所述第二像素值是否关于所述目标像素值对称，进而所述第二像素值是否呈阶梯状递增，在所述目标像素值大于所述第一像素值并小于所述第二像素值，且所述第一像素值和所述第二像素值关于所述目标像素值对称，所述第二像素值呈阶梯状递增时，将所述目标像素值对应的所述中心点作为所述目标重叠区域像素点。可以理解的是，所述中心点满足以下条件：
P
i＝max{P
1，P
2，...Pn}；Q
i＝max{Q
1，Q
2，...Q
n}；
P
1≈P
n，P
2≈P
n-1，...P
i≈P
n/2，Q
1≈Q
n，Q
2≈Q
n-1，...Q
i≈Q
n/2；
Q
i-Q
i-1＝c(i+1≤n/2),Q
i-Q
i+1＝-c(i≥n/2)。
其中，所述Pi和Qi为所述中心点的目标像素值，所述P1，P2，...Pn-1，Pn为所述第一其他像素点的第一像素值，所述Q1，Q2，...Qn-1，Qn为所述第二其他像素点的第二像素值，c为常数。
在一实施方式中，在确定所述中心点满足上述条件后，将所述中心点作为所述目标重叠区域像素点，并将所述延伸线确定为目标延伸线，可以理解的是，所述目标延伸线以目标重叠区域像素点为中心点，并且所述目标延伸线除所述目标重叠区域像素点以外的其他像素点的像素值比所述目标重叠区域像素点的像素值低，并且所述其他像素点的像素值关于所述目标重叠区域像素点的像素值对称。
在一实施方式中，在确定所述目标重叠区域像素点后，根据所述目标重叠区域像素点确定所述鞍点以及所述目标方向，参照图8，所述步骤S24包括：
步骤S241，获取所述目标延伸线与所述起始位置之间的目标夹角，将所述目标夹角确定为所述目标延伸线的方位角；
步骤S242，将所述目标重叠区域像素点确定为所述鞍点，并根据所述方位角确定所述目标方向。
在一实施方式中，所述目标延伸线为所述目标重叠区域像素点对应的延伸线中的目标延伸线，所述目标重叠区域像素点为所述目标延伸线上的极大值像素点和所述目标延伸线对应的目标垂线上的极小值像素点，所述目标延伸线上除所述目标重叠区域像素点以外的其他像素点对应的像素值比所述目标重叠区域像素点的像素值低，并关于所述目标重叠区域像素点的像素值对 称。
在一实施方式中，在所述起始位置为x轴时，获取所述目标延伸线与x轴的夹角，将所述夹角确定为所述目标延伸线的方位角，进而将所述方位角确定为所述目标方向对应的方位角，以确定所述目标方向。
在一实施方式中，在获取所述目标重叠区域像素点后，获取所述目标重叠区域像素点所在位置，并将所述目标重叠区域像素点所在位置确定为所述鞍点所在位置。
在本申请实施例中，通过以重叠区域像素点为中心，并以预设位置作起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线，进而根据所述灰度山图中各个像素点的像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点，进而根据所述目标重叠区域像素点中确定目标延伸线，进而将所述目标重叠区域像素点所在位置确定为所述鞍点所在位置，将所述目标延伸线对应的方位角确定为所述目标方向对应的方位角，以确定所述目标方向，通过以重叠区域像素点为中心构建多条延伸线，进而判断所述重叠区域像素点是否为目标重叠区域像素点，进而确定所述鞍点以及所述目标方向，进而以所述鞍点以及所述目标方向确定分割线，以根据分割线分割所述重叠区域，这样便可以根据分割线对重叠区域进行分割，得到单个矿石的完整图像，解决了传统的分割算法因矿石图像轮廓和灰度的不规则性导致图像过分割的问题，提高了矿石图像分割的准确性。
在一实施方式中，基于上述所有实施例，参照图9，所述重叠区域包括至少一个鞍点，所述步骤S30包括：
步骤S31，根据各个所述鞍点的位置信息获取各个所述鞍点之间的距离以及各个所述鞍点对应的目标方向对应的方位角之间的方位角差值；
步骤S32，确定所述鞍点之间的距离小于或等于预设距离阈值，且所述方位角差值小于或等于预设角度阈值，根据各个所述鞍点的位置信息计算所有所述鞍点的坐标均值以及根据所述目标方向的方位角计算得出方位角均值；
步骤S33，将所述坐标均值所在位置确定为所述分割线对应的分割点，根据所述方位角均值确定所述分割线的分割方向；
步骤S34，基于所述分割点以及所述分割方向确定分割线。
在本申请实施例中，在所述重叠区域可能会存在多个鞍点的情况，基于多个鞍点处于同一个重叠区域，各个鞍点之间的距离较近，若基于各个鞍点 均确定一条分割线，容易导致产生过多的分割线，而在实际的分割矿石图像过程中，若各个鞍点的距离较近时，只需用一条分割线即可分割所述重叠区域，基于此，本申请提出了一种确定分割线的方法。
在一实施方式中，在获取各个鞍点后，根据所述鞍点的位置信息获取各个所述鞍点之间的距离以及根据所述鞍点对应的目标方向对应的方位角差值，所述位置信息为所述鞍点对应的横坐标值以及纵坐标值，例如，在重叠区域A存在鞍点A，鞍点B，鞍点C，鞍点A对应的目标方向的方位角为G1，鞍点B对应的目标方向对应的方位角为G2，鞍点C对应的目标方向对应的方位角为G3，获取鞍点A与所述鞍点B之间的距离L1，鞍点A与鞍点C之间的距离L2，鞍点B与鞍点C之间的距离L3，以及方位角差值为：G1-G2，G2-G3，G1-G3。
在一实施方式中，在获取各个所述鞍点之间的距离以及各个所述鞍点对应的目标方向对应的方位角差值后，判断所述距离是否小于等于预设距离预设以及判断所述方位角差值是否小于或等于预设角度阈值，在所述鞍点之间的距离小于或等于预设距离阈值，且所述方位角差值小于或等于预设角度阈值，根据所述鞍点的位置信息获取所有所述鞍点对应的坐标均值以及根据所述目标方向获取所述鞍点对应的方位角均值，所述坐标均值包括横坐标均值以及纵坐标均值，例如：在重叠区域A存在鞍点A，鞍点B，鞍点C，鞍点A的位置信息为(x1，y1)以及对应的目标方向的方位角为G1，鞍点B的位置信息为(x2，y2)以及对应的目标方向对应的方位角为G2，鞍点C的位置信息为(x2，y3)以及对应的目标方向对应的方位角为G3，即所述横坐标均值为(x1+x2+x2)/3，所述纵坐标均值为(y1+y2+y2)/3，进而根据所述横坐标均值以及所述纵坐标均值确定坐标均值为((x1+x2+x2)/3，(y1+y2+y2)/3)，所述方位角均值为(G1+G2+G3)/3。
在一实施方式中，在又一实施例中，所述灰度山图可能包括多个重叠区域，各个重叠区域可能包括多个鞍点，本申请实施例在获取所述灰度山图各个鞍点之后，将所述鞍点分成多个鞍点集合，各个鞍点集合里面包括至少两个鞍点，所述将所述鞍点分成多个鞍点集合的方式为获取各个鞍点两两之间的距离以及方位角差值，响应于鞍点之间的距离小于或等于预设距离阈值和方位角差值都小于或等于预设角度阈值的情况，将距离小于或等于预设距离阈值和方位角差值都小于或等于预设角度阈值确定为同一个鞍点集合中的鞍点，可以理解的是，同一个鞍点集合中的鞍点两两之间的距离均小于或等于 所述预设距离阈值，且同一个鞍点集合中的鞍点对应的目标方向对应的方位角两两之间的方位角差值均小于等于预设角度阈值。
在一实施方式中，在获取各个所述鞍点集合后，根据各个所述鞍点集合中的鞍点确定所述鞍点集合对应的坐标均值以及方位角均值，不同的鞍点集合对应的坐标均值不同以及不同的鞍点集合对应的方位角均值可以相同，也可以不同。
在一实施方式中，在获取所述坐标均值以及所述方位角均值后，将所述坐标均值所在位置确定为所述分割线对应的分割点，并根据所述方位角均值确定所述分割线的分割方向，进而基于所述分割点以及所述分割方向确定所述分割线，进而根据所述分割线对所述重叠区域进行分割。
在本申请实施例中，在获取各个鞍点后，获取各个所述鞍点两两之间的距离以及各个所述鞍点对应的目标方向对应的方位角差值，在所述距离小于或等于预设距离阈值，且方位角差值小于或等于预设角度阈值时，获取所述距离小于或等于预设距离阈值且方位角差值小于或等于预设角度阈值的至少一个鞍点，进而获取所述至少一个鞍点对应的坐标均值以及方位角均值，进而根据所述坐标均值以及所述方位角均值确定分割线，以避免鞍点之间的距离过近时，生成多条分割线的问题，从而减少了计算量，进而提高了矿石图像的分割效率。
此外，本申请实施例还提出一种计算机可读存储介质，所述计算机可读存储介质上存储有矿石图像的分割程序，所述矿石图像的分割程序被处理器执行时实现如上所述的各个实施例的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者系统不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者系统所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者系统中还存在另外的相同要素。
上述本申请实施例序号仅仅为了描述，不代表实施例的优劣。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体 现出来，该计算机软件产品存储在如上所述的一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端设备(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
以上仅为本申请的可选实施例，并非因此限制本申请的专利范围，凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换，或直接或间接运用在其他相关的技术领域，均同理包括在本申请的专利保护范围内。
Claims (9)
- 一种矿石图像的分割方法，其中，所述矿石图像的分割方法的步骤包括：获取矿石的灰度山图；根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向，所述鞍点为所述目标方向上的极大值像素点和所述目标方向对应的垂直方向上的极小值像素点的集合；根据所述鞍点的位置信息以及所述目标方向确定分割线，并以所述分割线分割所述矿石。
- 如权利要求1所述的矿石图像的分割方法，其中，所述获取矿石的灰度山图的步骤包括：接收摄像头采集的矿石图像；将所述矿石图像转换成所述灰度山图。
- 如权利要求1所述的矿石图像的分割方法，其中，所述根据所述灰度山图中各个像素点的像素值确定重叠区域的鞍点以及所述鞍点对应的目标方向的步骤包括：根据所述灰度山图中的各个像素点的像素值获取重叠区域像素点；以任一重叠区域像素点为中心点并以预设位置作起始位置，沿预设方向构建多条夹角为预设角度的延伸线以及所述延伸线对应的垂线；根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线；根据所述目标重叠区域像素点确定所述鞍点以及所述目标方向。
- 如权利要求3所述的矿石图像的分割方法，其中，所述根据所述像素值、所述延伸线以及所述垂线获取所述重叠区域像素点中的目标重叠区域像素点以及所述延伸线中的目标延伸线的步骤包括：根据所述像素值获取所述延伸线上除所述中心点以外的第一其他像素点的第一像素值、所述垂线上除所述中心点以外的第二其他像素点的第二像素值，以及所述中心点的目标像素值；在所述目标像素值大于所述第一像素值并小于所述第二像素值第一像素值，且所述第一像素值和所述第二像素值关于所述目标像素值对称，所述第二像素值呈阶梯状递增时，将所述目标像素值对应的所述中心点作为所述目标重叠区域像素点，并将所述延伸线确定为目标延伸线。
- 如权利要求3所述的矿石图像的分割方法，其中，所述根据所述目标重叠区域像素点确定鞍点以及所述目标方向的步骤包括：获取所述目标延伸线与所述起始位置之间的目标夹角，将所述目标夹角确定为所述目标延伸线的方位角；将所述目标重叠区域像素点确定为所述鞍点，并根据所述方位角确定所述目标方向。
- 如权利要求1所述的矿石图像的分割方法，其中，所述重叠区域包括至少一个鞍点，所述根据所述鞍点的位置信息以及所述目标方向确定分割线的步骤包括：根据各个所述鞍点的位置信息获取各个所述鞍点之间的距离以及各个所述鞍点对应的目标方向对应的方位角之间的方位角差值；确定所述鞍点之间的距离小于或等于预设距离阈值，且所述方位角差值小于或等于预设角度阈值，根据各个所述鞍点的位置信息计算所有所述鞍点的坐标均值以及根据所述目标方向的方位角计算得出方位角均值；将所述坐标均值所在位置确定为所述分割线对应的分割点，根据所述方位角均值确定所述分割线的分割方向；基于所述分割点以及所述分割方向确定分割线。
- 如权利要求2所述的矿石图像的分割方法，其中，所述将所述矿石图像转换成所述灰度山图的步骤包括：对所述矿石图像进行预处理，以获取所述矿石图像对应的二值化图像，所述预处理的方式包括二值化处理和/或去噪声处理的至少一种；将所述二值化图像进行多次腐蚀处理，获取每次腐蚀处理后的所述二值化图像中各个像素点的像素值；依次将各个所述像素值进行叠加，以形成叠加后的所述二值化图像；将所述叠加后的所述二值化图像确定为所述灰度山图。
- 一种矿石图像的分割装置，其中，所述矿石图像的分割装置包括：存储器、处理器及存储在所述存储器上并可在所述处理器上运行的矿石图像的分割程序，所述矿石图像的分割程序被所述处理器执行时实现如权利要求1至7中任一项所述的矿石图像的分割方法的步骤。
- 一种计算机可读存储介质，其中，所述计算机可读存储介质上存储有矿石图像的分割程序，所述矿石图像的分割程序被处理器执行时实现如权利要求1至7中任一项所述的矿石图像的分割方法的步骤。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111399105.5 | 2021-11-23 | ||
| CN202111399105.5A CN114283155B (zh) | 2021-11-23 | 2021-11-23 | 矿石图像的分割方法、装置及计算机可读存储介质 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023092965A1 true WO2023092965A1 (zh) | 2023-06-01 |
Family
ID=80869859
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/094213 WO2023092965A1 (zh) | 2021-11-23 | 2022-05-20 | 矿石图像的分割方法、装置及计算机可读存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114283155B (zh) |
| WO (1) | WO2023092965A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823827A (zh) * | 2023-08-29 | 2023-09-29 | 山东德信微粉有限公司 | 一种基于图像处理的矿石粉碎效果评价方法 |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114283155B (zh) * | 2021-11-23 | 2023-07-04 | 赣州好朋友科技有限公司 | 矿石图像的分割方法、装置及计算机可读存储介质 |
| CN117576135A (zh) * | 2023-11-27 | 2024-02-20 | 北京霍里思特科技有限公司 | 一种基于矿石图像分割矿石的方法、设备和存储介质 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080123958A1 (en) * | 2006-06-26 | 2008-05-29 | Ratner Edward R | Computer-implemented method for efficient image segmentation using automated saddle-point detection |
| CN101710424A (zh) * | 2009-12-22 | 2010-05-19 | 中国矿业大学(北京) | 矿石图像分割方法 |
| CN105096283A (zh) * | 2014-04-29 | 2015-11-25 | 华为技术有限公司 | 全景图像的获取方法和装置 |
| CN114283155A (zh) * | 2021-11-23 | 2022-04-05 | 赣州好朋友科技有限公司 | 矿石图像的分割方法、装置及计算机可读存储介质 |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP6907767B2 (ja) * | 2017-07-06 | 2021-07-21 | 富士通株式会社 | 磁性体シミュレーション装置、磁性体シミュレーションプログラム、及び磁性体シミュレーション方法 |
| CN110853047B (zh) * | 2019-10-12 | 2023-09-15 | 平安科技（深圳）有限公司 | 智能图像分割及分类方法、装置及计算机可读存储介质 |
| CN111047555B (zh) * | 2019-11-13 | 2023-10-17 | 鞍钢集团矿业有限公司 | 基于图像处理技术的矿石图像粒度检测算法 |
-
2021
- 2021-11-23 CN CN202111399105.5A patent/CN114283155B/zh active Active
-
2022
- 2022-05-20 WO PCT/CN2022/094213 patent/WO2023092965A1/zh unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080123958A1 (en) * | 2006-06-26 | 2008-05-29 | Ratner Edward R | Computer-implemented method for efficient image segmentation using automated saddle-point detection |
| CN101710424A (zh) * | 2009-12-22 | 2010-05-19 | 中国矿业大学(北京) | 矿石图像分割方法 |
| CN105096283A (zh) * | 2014-04-29 | 2015-11-25 | 华为技术有限公司 | 全景图像的获取方法和装置 |
| CN114283155A (zh) * | 2021-11-23 | 2022-04-05 | 赣州好朋友科技有限公司 | 矿石图像的分割方法、装置及计算机可读存储介质 |
Non-Patent Citations (3)
| Title |
|---|
| KEN CHEN ; YICONG WANG ; GANGYI JIANG ; LARRY E. BANTA: "Saddle point detection for connecting objects in 2D images based on mathematic programming restraints", PROGRESS IN INFORMATICS AND COMPUTING (PIC), 2010 IEEE INTERNATIONAL CONFERENCE ON, IEEE, PISCATAWAY, NJ, USA, 10 December 2010 (2010-12-10), Piscataway, NJ, USA , pages 786 - 790, XP031849589, ISBN: 978-1-4244-6788-4 * |
| WANG YICONG, CHEN KEN, ZHAO PAN: "Programming Restraint Based Saddle-point Identification for Granule Images", JOURNAL OF IMAGE AND GRAPHICS, ZHONGGUO TUXIANG TUXING XUEHUI, CN, vol. 15, no. 3, 31 March 2010 (2010-03-31), CN , pages 385 - 391, XP093068732, ISSN: 1006-8961 * |
| WANG YI-CONG, CHEN KEN: "Image segmentation method of touching objects based on saddle-point search", COMPUTER ENGINEERING, SHANGHAI JISUANJI XUEHUI, CN, vol. 35, no. 10, 20 May 2009 (2009-05-20), CN , pages 203 - 205, XP093068735, ISSN: 1000-3428 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116823827A (zh) * | 2023-08-29 | 2023-09-29 | 山东德信微粉有限公司 | 一种基于图像处理的矿石粉碎效果评价方法 |
| CN116823827B (zh) * | 2023-08-29 | 2023-11-10 | 山东德信微粉有限公司 | 一种基于图像处理的矿石粉碎效果评价方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114283155B (zh) | 2023-07-04 |
| CN114283155A (zh) | 2022-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2023092965A1 (zh) | 矿石图像的分割方法、装置及计算机可读存储介质 | |
| CN108009543B (zh) | 一种车牌识别方法及装置 | |
| CN108304775B (zh) | 遥感图像识别方法、装置、存储介质以及电子设备 | |
| CN108875723B (zh) | 对象检测方法、装置和系统及存储介质 | |
| WO2019041519A1 (zh) | 目标跟踪装置、方法及计算机可读存储介质 | |
| US20160343165A1 (en) | Method for displaying augmented reality content based on 3d point cloud recognition, and apparatus and system for executing the method | |
| US9384398B2 (en) | Method and apparatus for roof type classification and reconstruction based on two dimensional aerial images | |
| CN108986152B (zh) | 一种基于差分图像的异物检测方法及装置 | |
| CN110443258B (zh) | 文字检测方法、装置、电子设备及存储介质 | |
| US9058655B2 (en) | Region of interest based image registration | |
| CN112381183B (zh) | 目标检测方法、装置、电子设备及存储介质 | |
| CN110431563B (zh) | 图像校正的方法和装置 | |
| CN109948521B (zh) | 图像纠偏方法和装置、设备及存储介质 | |
| Chung et al. | Adaptive cloud offloading of augmented reality applications on smart devices for minimum energy consumption | |
| CN108052869B (zh) | 车道线识别方法、装置及计算机可读存储介质 | |
| CN114283435A (zh) | 表格提取方法、装置、计算机设备、存储介质 | |
| WO2022062853A1 (zh) | 遥感图像的配准方法、装置、设备、存储介质及系统 | |
| CN109389628B (zh) | 图像配准方法、设备和存储介质 | |
| CN111191619A (zh) | 车道线虚线段的检测方法、装置、设备和可读存储介质 | |
| US10354409B2 (en) | Image processing device, image processing method, and non-transitory computer-readable recording medium | |
| CN112053360A (zh) | 图像分割方法、装置、计算机设备及存储介质 | |
| CN111639639B (zh) | 检测文本区域的方法、装置、设备及存储介质 | |
| CN110619597A (zh) | 一种半透明水印去除方法、装置、电子设备及存储介质 | |
| CN114022658A (zh) | 一种目标检测方法、装置、存储介质及终端 | |
| CN107678655B (zh) | 一种图像要素提取方法及图像要素提取系统 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22897064 Country of ref document: EP Kind code of ref document: A1 |