Mean filtering method and system
By constructing a mean filtering method of arbitrary shape, the problem of the boundary pixels of complex shaped regions in the image being affected by external factors in the existing technology is solved, realizing flexible detection and efficient defect recognition, and reducing missed detections and false detections.
Patent Information
- Application Number
- CN202111524209.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-12-14
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-12-14
AI Technical Summary
When processing images of complex shapes, existing technologies are prone to the influence of external pixels on boundary pixels, leading to missed and false detections of defects. Furthermore, they can only process rectangular or full-image areas and cannot flexibly set the detection area.
A mean filtering method is constructed by establishing a detection region of arbitrary shape, traversing pixels and calculating the mean, avoiding the influence of external factors on boundary pixels, and using a rectangular detection box to traverse and accumulate pixel gray values to obtain the final mean map.
It enables flexible detection of regions of arbitrary shape, with boundary pixels unaffected by external factors, thus improving the accuracy of defect detection and reducing missed and false detections.
Smart Images

Figure CN114399447B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of industrial automation machine vision technology, and in particular to a mean filtering method and system. Background Art
[0002] In the field of industrial automation machine vision, various types of defect detection projects are often encountered, and the mean filter method is widely used in the pre-processing steps of such projects. At present, many methods on the market process the entire image or intercept the rectangular area image, and process it by filling pixels when processing the image boundary. For example, the mean filter in the opencv open source library for image processing commonly used in the field of machine vision is like this. The disadvantages of this processing are as follows: First, the processing area can only be a rectangle or the entire image, and it is not effective for processing complex arbitrary shape area images. Second, because the image boundary is processed by filling pixels, when processing the image boundary pixels, it will be affected by the pixels outside the image or outside the area, and the boundary pixels will be processed as defects, which makes it impossible to distinguish from the actual defects on the boundary, resulting in missed detection and false detection of products. Summary of the invention
[0003] The technical problem to be solved by the present invention is to provide a mean filtering method and system for at least one defect in the prior art.
[0004] The technical solution adopted by the present invention to solve the technical problem is: constructing a mean filtering method, comprising the following steps:
[0005] S1: Get the detection area image R, filter kernel width width and filter kernel height height;
[0006] S2: Traversing the pixel points of the area to be detected in the detection area image R;
[0007] S3: Taking the current pixel point as the anchor point, establish a rectangular detection frame with a filter kernel width width and a filter kernel height height, traverse the pixel coordinates in the rectangular detection frame in each row and column, and determine whether the pixel coordinates are within the area to be detected. If so, execute step S4; if not, traverse the next pixel coordinate in the rectangular detection frame; after traversing the rectangular detection frame of a pixel point, execute step S5;
[0008] S4: Accumulate the pixel grayscale value and the number of points of the pixel coordinates to obtain the total pixel grayscale value sum and the total number of points number;
[0009] S5: Divide the total pixel grayscale value by the total number of points to obtain the mean, and continue traversing until all the pixel points in the area to be detected in the detection area image R are traversed to obtain a final mean map M.
[0010] Preferably, in the mean filtering method described in the present invention, the step S1 further comprises obtaining an original image G to be processed;
[0011] Obtain the detection area image R, including:
[0012] An area to be detected of any position and any shape is determined in the original image G, and the pixel grayscale value of the area to be detected is converted to a preset value, and the pixel grayscale values of other areas are converted to 0, so as to obtain the detection area image R.
[0013] Preferably, in the mean filtering method described in the present invention, step S2 comprises:
[0014] S21: Calculate the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R, and intercept it in the detection area image R;
[0015] S22: traverse the pixel points in the rectangular area to determine whether the pixel point is in the area to be detected. If so, execute step S3; if not, traverse the next pixel point.
[0016] Preferably, in the mean filtering method of the present invention, before step S21, the method further includes:
[0017] S20: Determine whether the proportion of the area to be detected in the detection area image R within the size of the detection area image R is less than a threshold. If so, record the pixel points of the area to be detected and execute steps S3, S4 and S5; if not, execute step S21.
[0018] Preferably, in the mean filtering method of the present invention, intercepting within the detection area image R comprises:
[0019] The rectangular area is cut out in the detection area image R according to the rectangular area and using the original image G as a coordinate reference.
[0020] Preferably, in the mean filtering method of the present invention, judging whether the pixel point or the pixel coordinate is within the area to be detected includes:
[0021] Determine whether the pixel grayscale value of the pixel point or the pixel coordinate is a preset value.
[0022] Preferably, in the mean filtering method described in the present invention, step S3 further includes:
[0023] Determine whether the pixel coordinates exceed the rectangular area, and at the same time determine whether the pixel coordinates are within the area to be detected. If so, execute step S4; if not, traverse the next pixel coordinates in the rectangular detection frame.
[0024] Preferably, in the mean filtering method described in the present invention, step S3 further includes:
[0025] The left and right borders of the rectangular area are each expanded outward by (width-1) / 2 pixels, and the upper and lower borders are each expanded outward by (height-1) / 2 pixels.
[0026] Preferably, in the mean filtering method described in the present invention, the method further includes: S6: outputting a mean map M.
[0027] The present invention also constructs a mean filtering system, comprising:
[0028] An acquisition module is used to obtain the detection area image R, the filter kernel width width and the filter kernel height height;
[0029] A traversal module, used to traverse the pixel points of the area to be detected in the detection area image R;
[0030] The detection module is used to establish a rectangular detection frame with a filter kernel width width and a filter kernel height height with the current pixel point as an anchor point, traverse the pixel coordinates in the rectangular detection frame in each row and column, and determine whether the pixel coordinates are within the area to be detected. If so, the accumulation calculation module is executed; if not, the next pixel coordinates in the rectangular detection frame are traversed; after the rectangular detection frame of a pixel point is traversed, the mean calculation module is executed;
[0031] An accumulation calculation module, used for accumulating the pixel grayscale value and the number of points of the pixel coordinates to obtain a total pixel grayscale value sum and a total number of points number;
[0032] The mean calculation module is used to obtain the mean by dividing the total pixel grayscale value by the total number of points, and traverse all the pixel points in the area to be detected in the detection area image R to obtain the final mean map M.
[0033] By implementing the present invention, the following beneficial effects are achieved:
[0034] The present invention can flexibly set the area to be detected in any position and shape according to the actual detection needs of the user, and the pixels on the boundary of the area to be detected will not be affected by other images outside the area and outside the boundary. Regardless of whether the defect of the product is in the area to be detected or on the boundary, this method can have a good pre-processing effect. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:
[0036] Figure 1 It is a processing flow chart of mean filtering of the present invention;
[0037] Figure 2 The defect detection effect after pre-processing by this method, the length and width of the mean filter kernel are both 51, where (a) and (b) are the original color images, (c) is the original image converted to a grayscale image, (d) is the image after mean filtering, (e) is the difference image after subtracting the mean filter image from the original image, and (f) is the threshold segmentation image of the difference image;
[0038] Figure 3 This is the defect detection effect after pre-processing by the mean filtering method of the opencv image processing open source library. The length and width of the mean filter kernel are both 51. (a) and (b) are the original color images, (c) is the original image converted to a grayscale image, (d) is the image after mean filtering, (e) is the difference image after subtracting the mean filtering image from the original image, and (f) is the threshold segmentation image of the difference image;
[0039] Figure 4 This is the defect detection effect after pre-processing by the mean filtering method of the opencv image processing open source library. The length and width of the mean filter kernel are both 11. (a) and (b) are the original color images, (c) is the original image converted to a grayscale image, (d) is the image after mean filtering, (e) is the difference image after subtracting the mean filtering image from the original image, and (f) is the threshold segmentation image of the difference image;
[0040] Figure 5 It is a module block diagram of the mean filtering system of the present invention. DETAILED DESCRIPTION
[0041] In order to have a clearer understanding of the technical features, purposes and effects of the present invention, specific embodiments of the present invention are now described in detail with reference to the accompanying drawings.
[0042] It should be noted that the flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor do they have to be executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.
[0043] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.
[0044] In the field of industrial automation machine vision, the demand for product defect detection is relatively large and the scenarios are relatively complex. The detection area of the product may be arbitrary or variable. It is far from enough to use only rectangular detection areas for processing. Therefore, it is necessary to consider detection areas of any shape at any position as processing areas, such as Figure 2 (b) shows a common circuit board, in which only the defects in the white polygonal box area in the product image (hereinafter referred to as "actual detection area") need to be detected, and the defects are marked as two places with white circles. The general processing flow is to first convert the original color image into a grayscale image, then process the grayscale image through mean filtering to obtain a mean image, and then subtract the mean image from the grayscale image to obtain a difference image, and finally binarize the difference image to process the defects. In fact, only the difference in the mean filtering process has a greater impact on the final result, so you only need to optimize and improve the mean filtering method to achieve good results. If you use the mean filtering method of the image library opencv directly, you need to take the minimum circumscribed parallel rectangle of the actual detection area as the area of interest of the image, such as Figure 3 (b) The white rectangle is the detection area, which will also process areas outside the actual detection area, such as Figure 3 (d) is the image processed by the mean filtering method. It can be seen that the areas outside the actual detection area are also blurred, which is unnecessary. Figure 3 As shown in (e) and (f), although the OpenCV method can process the defects in the actual detection area, it will be affected by the pixels outside the actual detection area when processing the boundary pixels, causing pixels with the same grayscale as the defects to appear around the boundary. These pixels will be mistaken for defects, and these pixels will be connected to the real defects on the boundary, resulting in missed detection and false detection of the product. No matter how large or small the filter kernel size used by the OpenCV mean filter is, there is no good effect, such as Figure 3 As shown in the figure, the effect of setting a large filter kernel is shown. Although the defects are detected, there are too many boundary interferences, such as Figure 4 As shown in the figure, the effect of setting a small filter kernel is shown. Although the boundary effect is reduced, the real defects cannot be processed, and the actual defects are connected with the boundary interference, resulting in the inability to extract the actual defects. Taking the above factors into consideration, the present invention designs a mean filtering method, such as Figure 2As shown in the figure, this method only needs to process the actual detection area and will not be affected by pixels outside the image and the actual detection area. It can effectively detect defects. The advantages of this method are as follows:
[0045] 1. Users can set the detection area of any shape and position according to the detection requirements, unlike the traditional detection method which can only detect rectangular areas or full-image detection.
[0046] 2. The pixels on the boundary of the detection area will not be affected by other images outside the detection area and outside the boundary. The defects on the boundary can be found accurately without causing false detection or missed detection of boundary pixels.
[0047] 3. It has a better detection effect for large or small defects in the product.
[0048] Specifically, Figure 1 As shown, the present invention discloses a mean filtering method, which is particularly prominent in industrial defect detection, such as electronic circuit board detection, screen scratch detection, mobile phone cover scratch detection, automotive power battery pit detection, tile crack detection, welding quality detection, dispensing quality detection, etc. The method includes the following steps:
[0049] Step S1: Obtain the detection area image R, filter kernel width width and filter kernel height height through external input;
[0050] Step S2: traverse the pixel points of the area to be detected in the detection area image R;
[0051] Step S3: With the current pixel point as the anchor point (horizontal coordinate r, vertical coordinate c), a rectangular detection frame of filter kernel width width and filter kernel height height is established, and the pixel coordinates (x, y) in the rectangular detection frame are traversed in each row and column to determine whether the pixel coordinates are within the area to be detected. If so, execute step S4; if not, traverse the next pixel coordinate in the rectangular detection frame; after traversing the rectangular detection frame of a pixel point, execute step S5.
[0052] Wherein, determining whether the pixel coordinates are within the area to be detected includes: determining whether the pixel grayscale value of the pixel coordinates is a preset value. In some embodiments, the preset value is 255, and in some other embodiments, the preset value may not be limited to 255.
[0053] Step S4: Accumulate the pixel grayscale value and the number of points of the pixel coordinates, as shown in formula (1) and formula (2), to obtain the total pixel grayscale value sum and the total number of points number;
[0054]
[0055]
[0056] Step S5: If the accumulated number of points is 0, the mean of the grayscale value of the pixel is 0. Otherwise, the total pixel grayscale value is divided by the total number of points to get the mean, and the pixel points of the area to be detected in the detection area image R are traversed until the final mean map M is obtained, as shown in formula (3). The number of points is a variable with an initial value of 0. When traversing the detection box with a length of width and a width of height, if the current pixel coordinate point meets the requirements, the value of the point is increased by 1;
[0057]
[0058] Step S6: Output the mean map M.
[0059] Furthermore, in this embodiment, step S1 also includes obtaining the original color image G to be processed, that is, Figure 2 (a). The step of obtaining the detection area image R comprises:
[0060] Determine the area of any shape and position that needs to be detected in the original image G, that is, Figure 2 (b) and convert the pixel grayscale values of the area to be detected into a preset value, and convert the pixel grayscale values of other areas into 0, to obtain the detection area image R. Among them, the area to be detected in any position and any shape can be manually traced by the user for determination, or can be automatically identified by a computer. In some embodiments, the preset value is 255, 255 is required to be processed, and 0 is not required to be processed. In some other embodiments, the preset value may not be limited to 255.
[0061] Accordingly, step S2 includes:
[0062] Step S21: Calculate the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R, and intercept it in the detection area image R.
[0063] The interception in the detection area image R includes: intercepting a rectangular area in the detection area image R according to the rectangular area and taking the original image G as a coordinate reference.
[0064] Step S22: traverse the pixel points in the rectangular area to determine whether the pixel point is in the area to be detected. If so, execute step S3; if not, traverse the next pixel point.
[0065] Wherein, determining whether the pixel point is within the area to be detected includes: determining whether the pixel grayscale value of the pixel point is a preset value. In some embodiments, the preset value is 255, and in some other embodiments, the preset value may not be limited to 255.
[0066] In some embodiments, before step S21, the method further includes:
[0067] Step S20: Determine whether the proportion of the area to be detected in the detection area image R within the size of the detection area image R is less than a threshold. If so, record the pixel points of the area to be detected and execute steps S3, S4 and S5; if not, execute step S21. In some embodiments, the threshold may be 20%, and in other embodiments, the threshold may not be limited to 20%.
[0068] In some embodiments, step S3 further includes:
[0069] It is determined whether the pixel coordinates exceed the rectangular area, and at the same time whether the pixel coordinates are within the area to be detected. If so, step S4 is executed; if not, the next pixel coordinates in the rectangular detection frame are traversed.
[0070] In some other embodiments, in order to prevent the pixel coordinates from exceeding the rectangular area, step S3 further includes:
[0071] The left and right borders of the rectangular area are expanded outward by (width-1) / 2 pixels, and the upper and lower borders are expanded outward by (height-1) / 2 pixels. The advantage of this is that when traversing the pixel coordinates of the rectangular detection box of the current anchor point, there is no need to determine whether the pixel coordinates exceed the rectangular area.
[0072] like Figure 5 As shown, the present invention discloses a mean filtering system, comprising:
[0073] The acquisition module is used to obtain the detection area image R, the filter kernel width width and the filter kernel height height through external input;
[0074] A traversal module is used to traverse the pixel points of the area to be detected in the detection area image R;
[0075] The detection module is used to establish a rectangular detection frame with a filter kernel width width and a filter kernel height height with the current pixel point as the anchor point (horizontal coordinate r, vertical coordinate c), traverse the pixel coordinates (x, y) in the rectangular detection frame in each row and column, and determine whether the pixel coordinates are within the area to be detected. If so, the accumulation calculation module is executed; if not, the next pixel coordinate in the rectangular detection frame is traversed; after the rectangular detection frame of a pixel point is traversed, the mean calculation module is executed;
[0076] Wherein, determining whether the pixel coordinates are within the area to be detected includes: determining whether the pixel grayscale value of the pixel coordinates is a preset value. In some embodiments, the preset value is 255, and in some other embodiments, the preset value may not be limited to 255.
[0077] The accumulation calculation module is used to accumulate the pixel grayscale value and the number of points of the pixel coordinates, as shown in formula (1) and formula (2), to obtain the total pixel grayscale value sum and the total number of points number;
[0078]
[0079]
[0080] The mean calculation module is used to calculate the mean of the grayscale value of the pixel point if the accumulated number of points is 0, otherwise the total pixel grayscale value is divided by the total number of points to obtain the mean, and until all the pixels in the area to be detected in the detection area image R are traversed, the final mean map M is obtained, as shown in formula (3). The number of points is a variable with an initial value of 0. When traversing the detection box with a length of width and a width of height, if the current pixel coordinate point meets the requirements, the value of the point is increased by 1.
[0081]
[0082] The output module is used to output the mean map M.
[0083] Furthermore, in this embodiment, the obtaining module is also used to obtain the original color image G to be processed, that is, Figure 2 (a). The detection area image R is obtained, including:
[0084] Determine the area of any shape and position that needs to be detected in the original image G, that is, Figure 2 (b) and convert the pixel grayscale values of the area to be detected into a preset value, and convert the pixel grayscale values of other areas into 0, to obtain the detection area image R. Among them, the area to be detected in any position and any shape can be manually traced by the user for determination, or can be automatically identified by a computer. In some embodiments, the preset value is 255, 255 is required to be processed, and 0 is not required to be processed. In some other embodiments, the preset value may not be limited to 255.
[0085] Accordingly, the pixel points of the area to be detected in the detection area image R are traversed, including:
[0086] Calculate the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R, and intercept it in the detection area image R. Specifically, intercepting in the detection area image R includes: intercepting a rectangular area in the detection area image R according to the rectangular area and taking the original image G as a coordinate reference;
[0087] Traverse the pixel points in the rectangular area to determine whether the pixel point is in the area to be detected. If so, execute the detection module. If not, traverse the next pixel point. Specifically, determining whether the pixel point is in the area to be detected includes: determining whether the pixel grayscale value of the pixel point is a preset value. In some embodiments, the preset value is 255. In some other embodiments, the preset value may not be limited to 255.
[0088] In some embodiments, the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R is calculated and intercepted in the detection area image R, before which the following steps are further included:
[0089] It is determined whether the proportion of the area to be detected in the detection area image R within the size of the detection area image R is less than a threshold value. If so, the pixel points of the area to be detected are recorded, and the detection module, the cumulative calculation module and the mean calculation module are executed; if not, the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R is calculated, and the area is intercepted in the detection area image R. In some embodiments, the threshold value may be 20%, and in some other embodiments, the threshold value may not be limited to 20%.
[0090] In some embodiments, the detection module is also used to determine whether the pixel coordinates exceed the rectangular area, and at the same time determine whether the pixel coordinates are within the area to be detected. If so, the accumulation calculation module is executed; if not, the next pixel coordinate in the rectangular detection box is traversed.
[0091] In some other embodiments, in order to prevent the pixel coordinates from exceeding the rectangular area, the detection module is also used to expand the left and right boundaries of the rectangular area outward by (width-1) / 2 pixels and the upper and lower boundaries outward by (height-1) / 2 pixels. The advantage of this is that when traversing the pixel coordinates of the rectangular detection box of the current anchor point, there is no need to determine whether the pixel coordinates exceed the rectangular area.
[0092] By implementing the present invention, the following beneficial effects are achieved:
[0093] The present invention can flexibly set the area to be detected in any position and shape according to the actual detection needs of the user, and the pixels on the boundary of the area to be detected will not be affected by other images outside the area and outside the boundary. Regardless of whether the defect of the product is in the area to be detected or on the boundary, this method can have a good pre-processing effect.
[0094] It can be understood that the above embodiments only express the preferred implementation modes of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the patent scope of the present invention. It should be pointed out that, for ordinary technicians in this field, the above technical features can be freely combined without departing from the concept of the present invention, and several deformations and improvements can be made, which all belong to the protection scope of the present invention. Therefore, all equivalent changes and modifications made to the scope of the claims of the present invention should belong to the coverage of the claims of the present invention.
Claims
1. A mean filtering method, characterized in that: The following steps are involved: S1: Obtaining the original image G to be processed, the detection area image R, the filter kernel width width and the filter kernel height height; wherein obtaining the detection area image R comprises: determining an area to be detected of any position and any shape in the original image G, and converting the pixel grayscale value of the area to be detected to a preset value, and converting the pixel grayscale value of other areas to 0, to obtain the detection area image R; S2: Traversing the pixel points of the area to be detected in the detection area image R, including: S21: Calculate the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R, and intercept it in the detection area image R; S22: traverse the pixel points in the rectangular area to determine whether the pixel point is in the area to be detected. If so, execute step S3; if not, traverse the next pixel point; S3: Taking the current pixel point as the anchor point, establish a rectangular detection frame with a filter kernel width width and a filter kernel height height, traverse the pixel coordinates in the rectangular detection frame in each row and column, and determine whether the pixel coordinates are within the area to be detected. If so, execute step S4; if not, traverse the next pixel coordinate in the rectangular detection frame; after traversing the rectangular detection frame of a pixel point, execute step S5; S4: Accumulate the pixel grayscale value and the number of points of the pixel coordinate in the original image G to obtain the total pixel grayscale value sum and the total number of points number; S5: Divide the total pixel grayscale value by the total number of points to obtain the mean, and continue traversing until all the pixel points in the area to be detected in the detection area image R are traversed to obtain a final mean map M.
2. The mean filtering method according to claim 1, characterized in that: Before step S21, the following steps are also included: S20: Determine whether the proportion of the area to be detected in the detection area image R within the size of the detection area image R is less than a threshold. If so, record the pixel points of the area to be detected and execute steps S3, S4 and S5; if not, execute step S21.
3. The mean filtering method according to claim 1, characterized in that: The interception is performed within the detection area image R, comprising: The rectangular area is cut out in the detection area image R according to the rectangular area and using the original image G as a coordinate reference.
4. The mean filtering method according to claim 1, characterized in that: Determining whether the pixel point or the pixel coordinate is within the area to be detected includes: Determine whether the pixel grayscale value of the pixel point or the pixel coordinate in the detection area image R is a preset value.
5. The mean filtering method according to claim 1, characterized in that: The step S3 further comprises: Determine whether the pixel coordinates exceed the rectangular area, and at the same time determine whether the pixel coordinates are within the area to be detected. If so, execute step S4; if not, traverse the next pixel coordinates in the rectangular detection frame.
6. The mean filtering method according to claim 1, characterized in that: The step S3 further comprises: The left and right borders of the rectangular area are each expanded outward by (width-1) / 2 pixels, and the upper and lower borders are each expanded outward by (height-1) / 2 pixels.
7. The mean filtering method according to claim 1, characterized in that: The method further includes: S6: outputting a mean map M.
8. A mean filtering system, characterized in that: include: An acquisition module is used to obtain the original image G to be processed, the detection area image R, the filter kernel width width and the filter kernel height height; wherein obtaining the detection area image R includes: determining an area to be detected at any position and any shape in the original image G, and converting the pixel grayscale value of the area to be detected to a preset value, and converting the pixel grayscale value of other areas to 0, to obtain the detection area image R; A traversal module is used to traverse the pixel points of the area to be detected in the detection area image R, including: calculating the minimum circumscribed parallel rectangular area of the area to be detected in the detection area image R, and intercepting it in the detection area image R; traversing the pixel points in the rectangular area, judging whether the pixel point is in the area to be detected, if so, executing the detection module, if not, traversing the next pixel point; The detection module is used to establish a rectangular detection frame with a filter kernel width width and a filter kernel height height with the current pixel point as an anchor point, traverse the pixel coordinates in the rectangular detection frame in each row and column, and determine whether the pixel coordinates are within the area to be detected. If so, the accumulation calculation module is executed; if not, the next pixel coordinates in the rectangular detection frame are traversed; after the rectangular detection frame of a pixel point is traversed, the mean calculation module is executed; An accumulation calculation module, used for accumulating the pixel grayscale value and the number of points of the pixel coordinates in the original image G to obtain a total pixel grayscale value sum and a total number of points number; The mean calculation module is used to obtain the mean by dividing the total pixel grayscale value by the total number of points, and traverse all the pixel points in the area to be detected in the detection area image R to obtain the final mean map M.
Citation Information
Patent Citations
Round Mark point positioning method based on connected region filtering
CN103729632A
Image noise filtering method and system
CN105809630A
Method and device for processing image buffing, medium and electronic equipment
CN110378846A
Multistage median filtering method for smooth denoising of metallographic image
CN110992279A
Self-adaptive threshold segmentation method for grayscale image
CN113129326A
Cited By
Video noise reduction method based on hybrid filtering
CN120219228A
A video noise reduction method based on hybrid filtering
CN120219228B
