Multi-line laser light stripe center extraction method and device and object three-dimensional reconstruction method

Abstract

The invention provides a multi-line laser strip center extraction method and device, an object three-dimensional reconstruction method and system, a laser scanner and a computer readable storage medium, and the extraction method comprises the steps: carrying out the binarization processing of an original image, carrying out the analysis of a connected region, and obtaining a plurality of connected regions; determining a to-be-processed sub-image and a mask sub-image of each connected region, processing the to-be-processed sub-image through the mask sub-image, and determining a central point of each laser line segment by adopting a Gaussian weighted gray gravity center method, so as to obtain a central line of each laser line segment; according to the laser light strip center extraction method, the extraction precision of each center point can be improved, and the extraction efficiency is high.

Description

Technical Field

[0001] The present invention relates to the field of three-dimensional reconstruction technology, and in particular to a method and device for extracting the center of a multi-line laser light strip, a method and system for three-dimensional reconstruction of an object, a laser scanner, and a computer-readable storage medium. Background Art

[0002] With the development of laser technology, optical measurement has been widely used in many fields due to its high measurement accuracy, fast measurement speed and good adaptability, such as architecture, industrial design, cultural heritage protection, medical treatment, art and entertainment. Among them, scanners based on multi-line laser light strips have developed rapidly due to their strong anti-interference ability and no texture restrictions, especially handheld laser scanners, which are more popular as portable devices.

[0003] In a multi-line laser light strip scanner, the extraction of the center line of the multi-line laser light strip is very important, and the center point of the laser line segment involved in the extraction of the center line of the multi-line laser light strip is even more crucial. The existing center extraction algorithms for laser line segments include the geometric center method, the skeleton refinement method, the Steger algorithm, and the simple grayscale centroid algorithm. Among them, the geometric center method only extracts the geometric center of the light strip, and needs to obtain the geometric boundary of the light strip. It cannot meet the application of laser scanners, especially handheld laser scanners, in terms of calculating the center position, extraction accuracy, and extraction speed. Although the skeleton refinement method can quickly extract the center point of the structured light as the skeleton line from the skeleton to a single pixel, the center line extracted by this method is not complete, the loss is serious, and the accuracy is relatively poor. The Steger algorithm obtains the normal direction of the center point of the light strip by calculating the Hession matrix, and obtains the center point coordinates by finding the extreme value of the second-order Taylor polynomial of the grayscale distribution function. This algorithm has a large amount of calculation and low efficiency. The simple grayscale centroid algorithm cannot be used directly when multi-line laser light strips appear in the image, and it directly processes the entire image, which is also slow. Therefore, it is necessary to propose a method for extracting the center of a multi-line laser stripe that can be applied to a laser scanner. Summary of the invention

[0004] Based on the above situation, the main purpose of the present invention is to provide a method and device for extracting the center line of a multi-line laser light strip, a method and system for reconstructing a three-dimensional object, a laser scanner and a computer-readable storage medium, which can meet the needs of fast and high-precision extraction of the center line of a multi-line laser light strip in a laser scanner, thereby improving the efficiency of modeling and measurement.

[0005] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0006] A first aspect of the present invention provides a method for extracting the center of a multi-line laser stripe, comprising the steps of:

[0007] S10: performing binarization processing on the original image containing multiple laser light strips to obtain a binarized image; performing connected region analysis on the binarized image to obtain multiple connected regions;

[0008] S20: respectively determining the sub-images to be processed and the mask sub-images corresponding to the connected regions, and multiplying the pixel values ​​of the corresponding pixels in the sub-images to be processed and the mask sub-images to obtain filtered sub-images; respectively extracting the center points of the laser line segments in the filtered sub-images by using the Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segments in the filtered sub-images, that is, obtaining the center positions of the laser line segments in the original image;

[0009] in,

[0010] The sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area;

[0011] The mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

[0012] Preferably, in step S20, before determining the sub-image to be processed and the mask sub-image corresponding to each of the connected regions, the connected regions having a pixel number less than a preset threshold are first filtered out.

[0013] Preferably, the preset threshold is 10-20.

[0014] Preferably, in the step S20, before determining the to-be-processed sub-image and the mask sub-image corresponding to each of the connected regions, the method further includes: setting the region index value of each of the connected regions to i; wherein i is 1, 2, ..., N, and N is the number of connected regions;

[0015] Determining the mask sub-image corresponding to the connected region with the region index value i includes:

[0016] It is determined whether the region index value corresponding to each pixel in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area is i. If not, the pixel value of the pixel is set to 0, otherwise the pixel value of the pixel remains unchanged.

[0017] Preferably, in step S20, each of the connected regions is processed in parallel.

[0018] Preferably, in step S20, the center points of the laser line segments in the filter sub-images are extracted respectively using a Gaussian weighted grayscale centroid method, which specifically includes the steps of:

[0019] S251: determining a Gaussian weight matrix of the filter sub-image according to the following Gaussian weight function w(I(x, y)), and multiplying the Gaussian weight matrix by corresponding elements in the filter sub-image to obtain a first matrix;

[0020]

[0021] Where I(x,y) is the grayscale value of the pixel at the position (x,y);

[0022] S252: constructing a row window matrix and a column window matrix of the same size as the filtered sub-image, and multiplying the row window matrix and the column window matrix with corresponding elements in the first matrix respectively to obtain a second matrix and a third matrix with position weight and Gaussian weight information;

[0023] The element values ​​of each column in each row of the row window matrix are the same, which are the row position index values ​​of the filter sub-image corresponding to the row; the element values ​​of each row in each column of the column window matrix are the same, which are the column position index values ​​of the filter sub-image corresponding to the column;

[0024] S253: sum the second matrix and the first matrix by columns, and divide the sums of corresponding columns to obtain an initial column coordinate set of the laser center processed by row index as weight; sum the third matrix and the first matrix by rows, and divide the sums of corresponding rows to obtain an initial row coordinate set of the laser center processed by column index as weight.

[0025] S254: determining a preliminary center point set of the laser line segment in the filter sub-image according to the initial column coordinate set and the initial row coordinate set, wherein if the number of columns of the filter sub-image is greater than or equal to the number of rows, determining the preliminary center point set according to step S2541; if the number of rows of the filter sub-image is greater than the number of columns, determining the preliminary center point set according to step S2542;

[0026] S2541: round down each element value in the initial column coordinate set to obtain a first intermediate sequence set; average the corresponding element values ​​of the initial column coordinate set and the first intermediate sequence set respectively, as the row coordinates of each center point in the initial center point set; average each element value in the first intermediate sequence set and the element value in the initial row coordinate set with the element value as the position index, as the column coordinates of each center point in the initial center point set;

[0027] S2542: round down each element value in the initial row coordinate set to an integer to obtain a second intermediate sequence set, recorded as X={x1, x2, ..., xM}; average the corresponding element values ​​of the initial row coordinate set and the second intermediate sequence set respectively, as the column coordinates of each center point in the initial center point set; average each element value in the second intermediate sequence set and the element value in the initial column coordinate set with the element value as the position index, as the row coordinates of each center point in the initial center point set;

[0028] S255: Judge each center point in the initial center point set. For each center point in the initial center point set, select the four pixels above, below, left and right of the center point coordinates on the filter sub-image, and calculate the average grayscale value of the four pixels. If the average value is greater than the preset grayscale value, it is used as the confirmed center point of the laser line segment of the filter sub-image, otherwise the center point is discarded. Each confirmed center point is a plurality of centers of the laser line segment of the sub-image to be processed. Preset grayscale value Preferably, in step S20, after obtaining the center of the laser line segment in each filter sub-image, the center points of the same filter sub-image are sequentially connected along the size of their position index values ​​to obtain the center line of the corresponding connected area.

[0029] Preferably, in step S10, the original image is first subjected to Gaussian filtering or median filtering to obtain a preprocessed image, and then the preprocessed image is subjected to binarization processing;

[0030] In the step S20, the sub-image to be processed corresponding to each of the connected regions is the portion of the pre-processed image corresponding to the minimum circumscribed rectangle of the connected region.

[0031] A second aspect of the present invention provides a device for extracting the center of a multi-line laser light strip, comprising:

[0032] A connected region division unit is used to perform a binarization process on an original image containing multiple laser light strips to obtain a binary image; and to perform a connected region analysis on the binary image to obtain multiple connected regions;

[0033] Center extraction unit: used to respectively determine the sub-image to be processed and the mask sub-image corresponding to each of the connected areas, multiply the pixel values ​​of the corresponding sub-image to be processed with the corresponding pixel in the mask sub-image to obtain a filtered sub-image; extract the center points of the laser line segments in each of the filtered sub-images by Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segments in each of the filtered sub-images, that is, to obtain the center positions of each laser line segment in the original image;

[0034] in,

[0035] The sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area;

[0036] The mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

[0037] A third aspect of the present invention provides a method for three-dimensional reconstruction of an object, comprising the steps of:

[0038] S100: Acquire an original image of a multi-line laser light strip projected onto the object to be measured;

[0039] S200: extracting the center position of each laser line segment in the original image by using any of the extraction methods described above;

[0040] S300: Perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of the laser line segments.

[0041] A fourth aspect of the present invention provides a system for three-dimensional reconstruction of an object, comprising:

[0042] An acquisition module, used to acquire an original image of the multi-line laser light strips projected onto the object to be measured;

[0043] A centerline extraction module, used to extract the center position of each laser line segment in the original image by using any of the extraction methods described above;

[0044] The three-dimensional reconstruction module is used to perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of each of the laser line segments.

[0045] A fifth aspect of the present invention provides a laser scanner, comprising:

[0046] A multi-line laser, two of which are provided, for projecting a multi-line laser line toward the object to be measured;

[0047] An acquisition device is used to acquire images of multiple laser lines on the object to be measured to obtain an original image;

[0048] A processing device, comprising the above-mentioned object three-dimensional reconstruction system;

[0049] Wherein, the acquisition device is signal-connected to the processing device.

[0050] A sixth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the method for extracting the center of a multi-line laser light strip as described in any one of the above items, or implements the method for three-dimensional reconstruction of an object as described above.

[0051] The method for extracting the center of a multi-line laser light strip of the present invention converts the processing of the entire image into processing of local sub-images to be processed respectively, and the mask sub-image is also produced for a local area, therefore, the accuracy of mask production and the image processing accuracy of each local sub-image to be processed can be improved, and the center point of the processed filtered sub-image is extracted by combining the improved grayscale centroid method, i.e., the Gaussian weighted grayscale centroid method, so the extraction accuracy and speed of the center point are greatly improved, thereby improving the accuracy of the generated three-dimensional point cloud data and model, ensuring that the point cloud data is evenly distributed and has no obvious noise or deviation, so as to accurately capture the tiny details of the surface of the object to be measured, such as texture, convexity and curvature, etc., thereby improving the shape and characteristics of the object to be reconstructed in three dimensions, and facilitating the smooth progress of subsequent steps of generating three-dimensional point cloud data, such as registration, splicing, surface reconstruction, etc., thereby realizing fast and accurate three-dimensional modeling, and can be widely used in laser scanners, especially handheld laser scanners.

[0052] Other beneficial effects of the present invention will be explained in the specific implementation manner through the introduction of specific technical features and technical solutions. Through the introduction of these technical features and technical solutions, those skilled in the art should be able to understand the beneficial technical effects brought about by the technical features and technical solutions. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Preferred embodiments of the present invention will be described below with reference to the accompanying drawings.

[0054] Figure 1 A schematic flow chart of a preferred embodiment of a method for extracting the center line of a multi-line laser light strip provided by the present invention;

[0055] Figure 2 A binary image obtained by binarizing an original image in a specific embodiment of the method for extracting the center line of a multi-line laser light strip provided by the present invention;

[0056] Figure 3 For Figure 2 Schematic diagram of indexing and marking the original image in;

[0057] Figure 4 , Figure 5 Respectively Figure 2 The illustrated embodiment shows a mask area and an enlarged view of the mask area;

[0058] Figure 6 For Figure 5 The mask sub-image obtained after processing the mask area of ​​;

[0059] Figure 7 for Figure 6 The sub-image to be processed corresponding to the mask sub-image;

[0060] Figure 8 is the center position of the laser line segment obtained by processing the sub-image to be processed in the rectangular frame using steps S251 to S255;

[0061] Fig. 9 The center of the multi-line laser light strip is extracted by the extraction method of the present invention for the entire sub-image to be processed;

[0062] Fig.10 A schematic diagram of a specific embodiment of the laser scanner provided for the invention. DETAILED DESCRIPTION

[0063] The present invention is described below based on embodiments, but the present invention is not limited to these embodiments. In the following detailed description of the present invention, some specific details are described in detail. In order to avoid confusing the essence of the present invention, known methods, processes, procedures, and components are not described in detail.

[0064] In addition, persons of ordinary skill in the art will appreciate that the drawings provided herein are for illustration purposes and are not necessarily drawn to scale.

[0065] Unless the context clearly requires otherwise, throughout the specification and claims, the words "include", "comprising" and similar words should be interpreted in an inclusive sense rather than an exclusive or exhaustive sense; that is, in the sense of "including but not limited to".

[0066] In the description of the present invention, it should be understood that the terms "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying relative importance. In addition, in the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0067] In many technical fields such as construction and detection, a multi-line laser is used to emit a multi-line laser beam to the object to be tested, so that an original image with multi-line laser light strips can be collected on the object. The original image can be processed and used to construct a three-dimensional model of the object to be tested. The most important thing in the processing of the original image is to extract the center point of the multi-line laser light strips in the original image.

[0068] The present invention provides a method for extracting the center of a multi-line laser light strip, which can be applied to reverse engineering, medical assistance, defect detection, industrial measurement and positioning, cultural relics protection, entertainment and other fields. Specifically, it can be applied to electronic equipment in these fields, which can be used in conjunction with lasers, cameras, etc. to perform three-dimensional reconstruction of objects, such as scanners used in multiple fields, especially handheld laser scanners. Since the method for extracting the center of a multi-line laser light strip of the present invention has the advantages of high precision, rapidity and real-time performance, it can be better applied to the requirements of laser scanners for modeling, measurement and other aspects for the first time.

[0069] like Figure 1 As shown, the method for extracting the center of a multi-line laser light strip comprises the steps of:

[0070] S10: Binarization is performed on the original image containing multiple laser light strips to obtain a binary image, such as Figure 2 As shown; the connected region analysis is performed on the binary image to obtain multiple connected regions, such as Figure 3 As shown, each light strip is a connected area. The original image includes multiple laser line segments, and each connected area has only one continuous laser line segment. In this step, each laser line segment in the binary image forms a connected area.

[0071] S20: respectively determine the sub-image to be processed and the mask sub-image corresponding to each connected area, multiply the pixel values ​​of the corresponding pixels in the corresponding sub-image to be processed and the mask sub-image to obtain a filtered sub-image; respectively extract the center points of the laser line segments in each filtered sub-image by the Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segments in each filtered sub-image, that is, obtain the center positions of each laser line segment in the original image.

[0072] Among them, the sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area; the mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of the pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

[0073] That is, in step S20, the following steps are performed on each connected area:

[0074] S21: Determine the minimum circumscribed rectangle ROI of the current connected area;

[0075] S22: Determine the sub-image to be processed in the current connected region, that is, the portion of the original image corresponding to the minimum circumscribed rectangle ROI is used as the sub-image to be processed in the current connected region, such as Figure 7 As shown;

[0076] S23: Create a mask sub-image, that is, first use the portion of the binary image corresponding to the lowermost circumscribed rectangular ROI as the mask area (such as Figure 4 The rectangular frame, that is Figure 5 As shown), then the pixel values ​​of the pixels in the mask area outside the current connected area are set to 0, and the pixel values ​​of the remaining pixels remain unchanged. The mask area at this time is the mask sub-image corresponding to the current connected area (as shown Figure 6 shown);

[0077] S24: obtaining a filtered sub-image, that is, multiplying the pixel values ​​of the corresponding pixels in the sub-image to be processed and the mask sub-image as the pixel values ​​of the corresponding pixels in the filtered sub-image, thereby obtaining a filtered sub-image -;

[0078] S25: The center point of the laser line segment in the filter sub-image is extracted by using the Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segment in the filter sub-image, such as Figure 8 shown.

[0079] After steps S21 to S25 are performed on each connected area, the center point of each laser line segment in the entire original image is obtained, such as Fig. 9 shown.

[0080] The method for extracting the center of a multi-line laser light strip of the present invention first obtains a plurality of sub-images to be processed by connected region analysis, and produces a mask sub-image for each sub-image to be processed. Then, after the sub-image to be processed is processed by the mask sub-image, the center point of each laser line segment is determined by using the Gaussian weighted grayscale centroid method, thereby obtaining the center of each laser line segment. This method for extracting the center of a multi-line laser light strip converts the processing of the entire image into processing of local sub-images to be processed respectively, and the mask sub-image is also produced for a local area. Therefore, the accuracy of mask production and the image processing accuracy of each local sub-image to be processed can be improved, and the center point of the processed filtered sub-image is extracted by combining the improved grayscale centroid method, namely the Gaussian weighted grayscale centroid method. The extraction accuracy and speed of the center point are greatly improved, and the accuracy of the generated three-dimensional point cloud data and model is improved, ensuring that the point cloud data is evenly distributed and has no obvious noise or deviation, so as to accurately capture the tiny details of the surface of the object to be measured, such as the texture, convexity and curvature of the surface, thereby improving the shape and characteristics of the three-dimensionally reconstructed object, and facilitating the smooth progress of other processing steps after the generation of the three-dimensional point cloud data, such as model alignment, splicing, surface reconstruction, etc. It can be seen that the extraction method of the present invention can realize fast and accurate three-dimensional modeling and detection, and can be widely used in laser scanners, especially handheld laser scanners.

[0081] In step S10, the original image is first subjected to Gaussian filtering or median filtering, that is, the original image is preprocessed to obtain a preprocessed image, and then the preprocessed image is binarized. In this embodiment, in step S20, the sub-image to be processed corresponding to each connected area is the part of the preprocessed image corresponding to the minimum circumscribed rectangle of the connected area.

[0082] In an actual measurement environment, the laser beam emitted from a multi-line laser has a certain width in space, which is represented on the image as a laser light bar occupying a certain pixel area. In three-dimensional reconstruction, single pixel points or sub-pixel points are required to achieve accurate three-dimensional reconstruction. Due to the influence of laser speckle, the distribution of the laser light bar is not a more ideal light bar distribution, so the laser image is first preprocessed to remove noise while smoothing the protruding edges of the light bar. Specifically, Gaussian filtering or median filtering can be used, preferably Gaussian filtering, to perform Gaussian blur processing on the original image, and then change the laser cross-sectional light intensity distribution into a Gaussian distribution. The Gaussian kernel used for Gaussian filtering can be determined according to the image area occupied by the width of a single laser light bar in the original image. In the present invention, a Gaussian kernel of size 5*5 is used for Gaussian filtering, which can better remove Gaussian noise and smooth the protruding edges of the laser light bar.

[0083] In step S10, the original image is binarized. Specifically, a binarization threshold may be set, and the pixel values ​​of pixels in the original image (the preprocessed image in the embodiment where preprocessing is performed first) whose pixel values ​​are greater than the binarization threshold are set to 1, and the pixel values ​​of the remaining pixels are set to 0.

[0084] In the above embodiment, the specific method of performing connected region analysis on the binary image may adopt the connected region analysis method in the prior art, such as using a four-neighborhood or eight-neighborhood method to partition each pixel in the binary image.

[0085] Connectivity defines whether adjacent pixels are considered to be connected. Common connectivity includes 4-connectivity and 8-connectivity. Pixels connected in four directions, up, down, left, and right, are 4-connected. Pixels connected in eight directions, up, down, left, right, and diagonal, are 8-connected. After the original image is binarized, the image is scanned: traverse each pixel from top to bottom and from left to right, and use recursion or data structures such as queues / stacks to mark all pixels connected to the current pixel as the same area, with the label recorded as index i. When there are no pixel values ​​1 or 255 in the neighborhood, start looking for the next connected area. Continue searching and find that the current pixel value is 1 or 255 and is not marked, then start marking a new connected area with a label index of i+1.

[0086] When scanning an actual object, since the multi-line laser is modulated by the height of the object to be measured, the laser line may be broken in the original image. That is, when the same laser light bar projected by the multi-line laser is projected onto the object to be measured, the laser light bar in the collected original image may be multiple disconnected laser line segments (that is, multiple laser line segments on the original image correspond to the same laser light bar), or it may still be a continuous laser line. Therefore, the laser line segment in the original image can be part of the projection of the same laser light bar on the object to be measured, or it can be all.

[0087] Due to the modulation of the height of the object to be measured and the influence of environmental conditions such as stray light, in step S10, after the connected area analysis is performed, among the connected areas that may be obtained, some connected areas occupy a very small area (number of pixels), while some connected areas (number of pixels) have a larger area. For connected areas with particularly small areas, if step S20 is executed, even if these connected areas are valid connected areas (not obtained by environmental conditions such as stray light), at most a few points will be added to the finally determined center point, and the number of points can be completely ignored relative to the entire image, but it will greatly increase the processing efficiency of the entire image, and these connected areas may also be interference. In a preferred embodiment of the present invention, these connected areas will be removed. Specifically, in step S20, before determining the sub-images to be processed and the mask sub-images corresponding to each connected area, the connected areas with the number of pixels less than the preset threshold are first filtered out, that is, in step S20, the number of pixels occupied by each connected area obtained in step S10 will be first determined, and the connected areas with the number of pixels less than the preset threshold will be removed, and steps S21 to S25 will not be executed, thereby improving the efficiency of three-dimensional reconstruction and avoiding interference. Among them, the preset threshold can be determined based on the width of a single laser light strip of a multi-line laser light strip, etc., or it can be determined by experience. In the present invention, the preset threshold is preferably selected as 10 to 20, such as 10, 13, 15, 17, 19 or 20.

[0088] In step S20, the minimum circumscribed rectangle of the connected area can be determined specifically according to the coordinates of the starting point and the ending point of the connected area, such as taking the starting point and the ending point as the two endpoints of the diagonal of the rectangle respectively, taking the absolute value of the difference between the horizontal coordinates of the starting point and the ending point as the length of the rectangle, and taking the difference between the vertical coordinates as the width of the rectangle.

[0089] In a preferred embodiment, in order to facilitate the execution of steps S21 to S25, a region index value may be marked for each connected region first, that is, in step S20, before setting the sub-image to be processed and the mask sub-image corresponding to each connected region, the process further includes: setting the region index value of each connected region to i; wherein i is 1, 2, ... N, and N is the number of connected regions. In the embodiment of filtering out connected regions whose number of pixels is less than a preset threshold, the region index value may be set only for these retained connected regions.

[0090] In specific implementation, in some embodiments, a position index value may be set for each connected area synchronously when performing connected area analysis. In this case, the position index value may be directly used in step S20. When the connected areas whose pixel number is less than a preset threshold are first filtered out in step S20, the original area index value of the retained connected area may be directly used, or the area index value of the retained connected area may be reset.

[0091] In this embodiment, the region index values ​​of the pixels corresponding to the same connected region are the same, so the determination of the mask sub-image corresponding to the connected region with the region index value i in step S20 can be performed directly through the region index value, including: judging whether the region index value corresponding to each pixel in the part of the minimum circumscribed rectangle of the corresponding connected region on the binary image is i, if not, setting the pixel value of the pixel to 0, otherwise the pixel value of the pixel remains unchanged. That is to say, in the above step S23, the pixel values ​​of the pixels in the mask region with the region index value not being i are set to 0.

[0092] It can be seen that by setting the region index value for each connected region, different sub-image regions are obtained so as to be processed simultaneously, thereby improving the extraction efficiency of the center point of the laser line segment and the speed of 3D reconstruction. Of course, the generation of the mask sub-image can also be directly determined by judging whether the position of each pixel in the mask region is within the current connected region, and determining whether to set the pixel value of the pixel to 0.

[0093] In step S20, each connected region may be processed in parallel, that is, steps S21 to S25 may be performed on each connected region at the same time, so as to further improve the extraction efficiency of the line points in each laser line segment.

[0094] In another embodiment, in step S20, each connected region is processed serially, that is, after executing steps S21 to S25 on one of the connected regions, steps S21 to S25 are executed on another connected region until all connected regions are executed. Specifically, steps S21 to S25 can be executed on each connected region in order from the upper left to the lower right. In the embodiment where the region index value is set for each connected region, the steps can also be executed in order from small to large region index values. For example, when the region index values ​​are set to 1, 2, ..., N in sequence, step S20 includes the steps of:

[0095] S00: let i=1;

[0096] S02: Determine the sub-image to be processed and the mask sub-image corresponding to the connected area with the area index value i, multiply the pixel values ​​of the corresponding pixels in the sub-image to be processed and the mask sub-image to obtain a filtered sub-image; extract the center point of the laser line segment in the filtered sub-image by the Gaussian weighted grayscale centroid method, that is, perform steps S21 to S25 on the connected area with the area index value i;

[0097] S03: Determine whether i is less than N. If so, add 1 to i and return to step S02; otherwise, end. The obtained set of center points is the center position of each laser line segment in the original image.

[0098] In step S20, the center points of the laser line segments in the filter sub-images are extracted using the Gaussian weighted grayscale centroid method. Specifically, step S25 can use the following formulas (1) and (2) to extract the X coordinates x of the center points of the laser line segments one by one. centraid and the Y coordinate y centraid .

[0099]

[0100] Where I(x,y) is the grayscale value of the pixel at the position (x,y); w(I(x,y)) is the Gaussian weight function; M and N are the length and width of the filter sub-image, that is, the number of pixels in the X direction and the number of pixels in the Y direction of the filter sub-image, respectively.

[0101] In a preferred embodiment, when extracting the center points of the laser line segments in the filter sub-image, the X coordinates and Y coordinates of each center point of the laser line segment are extracted at one time using a matrix method. The center points of the laser line segments in the filter sub-image can be extracted using a Gaussian weighted grayscale centroid method in the row direction and the column direction, i.e., the initial column coordinate set and the initial row coordinate set described below. Then, multiple centers of the laser line segments in the filter sub-image are determined based on the two coordinate sets. Specifically, the above step S25 includes the steps

[0102] S251: According to the Gaussian weight function w(I(x,y)), i.e., the above formula (1), a Gaussian weight matrix of the filter sub-image is determined, and the Gaussian weight matrix is ​​multiplied by the corresponding elements in the filter sub-image to obtain a first matrix;

[0103] S252: constructing a row window matrix and a column window matrix of the same size as the filter sub-image, and multiplying the row window matrix and the column window matrix with corresponding elements in the first matrix, respectively, to obtain a second matrix and a third matrix with position weight and Gaussian weight information;

[0104] Among them, the element values ​​of each column in each row of the row window matrix are the same, which are the row position index values ​​of the corresponding filtered sub-images, and the element values ​​of each row in each column of the column window matrix are the same, which are the position index values ​​of the corresponding columns of the corresponding filtered sub-images; the position index value of the column position is used as the weight, and the values ​​of each column in the same column are the same; that is, in the row window matrix, the value of each element in each row is its row number, and in the column window matrix, the value of each element in each column is its column number, such as the coordinates of the upper left corner of the filtered sub-image are (i, j) (i.e., the corresponding original image The jth row, ith column), the coordinates of the lower right corner are (i+m, j+n) (that is, corresponding to the j+nth row and i+mth column of the original image), then the size of the row window matrix corresponding to the filtered sub-image is the same as the filtered sub-image, and the element values ​​of the first row are j, the element values ​​of the second row are j+1, and so on, the element values ​​of the nth row are j+n; the size of the column window matrix corresponding to the filtered sub-image is the same as the filtered sub-image, and the element values ​​of the first column are i, the element values ​​of the second column are i+1, and so on, the element values ​​of the mth row are j+m.

[0105] S253: Sum the columns of the second matrix and the first matrix respectively, and divide the sums of the corresponding columns to obtain the initial column coordinate set of the laser center processed by the row index as the weight, which is recorded as PointCol{A1, A2, ..., AN}; sum the rows of the third matrix and the first matrix respectively, and divide the sums of the corresponding rows to obtain the initial row coordinate set of the laser center processed by the column index as the weight. That is to say, the second matrix and the first matrix sum each column element respectively to obtain two sets of column summation data, and then divide the corresponding elements in the two sets of column summation data to obtain the initial column coordinate set, which is recorded as PointRow{B1, B2, ..., BM}; the third matrix and the first matrix sum each row element respectively to obtain two sets of row summation data, and then divide the corresponding elements in the two sets of row summation data to obtain the initial row coordinate set.

[0106] S254: determining the initial center point set of the laser line segment in the filter sub-image according to the initial column coordinate set and the initial row coordinate set, wherein if the number of columns of the filter sub-image is greater than or equal to the number of rows, determining the initial center point set of the laser line segment in the filter sub-image according to step S2541; if the number of rows of the filter sub-image is greater than the number of columns, determining the initial center point set of the laser line segment in the filter sub-image according to step S2542;

[0107] S2541: Round down each element value in the initial column coordinate set to obtain a first intermediate sequence set, recorded as Y={y1, y2, ..., yN}, average the corresponding element values ​​of the initial column coordinate set and the first intermediate sequence set, as the row coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image; average each element value in the first intermediate sequence set and the element value in the initial row coordinate set with the element value as the position index, as the column coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image. That is, the row coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image are (A1+y1) / 2, (A2+y2) / 2, ..., (AN+yN) / 2, respectively; the column coordinates are respectively the average value of y1 and the y1th element value in the initial row coordinate set, the average value of y2 and the y2th element value in the initial row coordinate set, ..., the average value of yN and the yNth element value in the initial row coordinate set.

[0108] S2542: round down each element value in the initial row coordinate set to obtain a second intermediate sequence set, recorded as X={x1, x2, ..., xM}, average the corresponding element values ​​of the initial row coordinate set and the second intermediate sequence set, respectively, as the column coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image; average each element value in the second intermediate sequence set and the element value in the initial column coordinate set with the element value as the position index, as the row coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image. That is, the column coordinates of each center point in the initial center point set of the laser line segment in the filter sub-image are (B1+x1) / 2, (B2+x2) / 2, ..., (BM+xM) / 2, respectively; and the row coordinates are the average value of x1 and the x1th element value in the initial column coordinate set, the average value of x2 and the x2th element value in the initial column coordinate set, ..., the average value of xM and the xMth element value in the initial column coordinate set.

[0109] Each center point in the initial center point set obtained in step S254 can be directly used as multiple centers of the laser line segment of the filter sub-image. In order to further improve the extraction accuracy of each center of the laser line segment of the filter sub-image, the following steps are further included after step S254:

[0110] S255: Judge each center point in the above-calculated preliminary center point set. For each center point in the preliminary center point set, select the top, bottom, left and right four pixels of the center point coordinates on the filtered sub-image, and calculate the average grayscale value of the four pixels. If the average is greater than the preset grayscale value, it is used as the confirmed center point of the laser line segment of the filtered sub-image, otherwise the center point is discarded. Each confirmed center point is the multiple centers of the laser line segment of the sub-image to be processed.

[0111] The laser center point coordinates extracted by the above method, x, y are sub-pixel coordinates, and meet the characteristics of the laser line in the handheld multi-line laser system being inclined along the diagonal line, providing accurate laser light strip center point coordinates for subsequent three-dimensional reconstruction.

[0112] After obtaining the center points of the laser colinear segments of each filter sub-image, in some embodiments, these center points are directly used for subsequent steps such as three-dimensional reconstruction or measurement. In some embodiments, in step S20, after obtaining the center of the laser line segment in each filter sub-image, the center points of the same filter sub-image are connected to obtain the center line of each laser line segment, and the center line is used to perform subsequent steps such as three-dimensional reconstruction or measurement. When connecting the center points of the same filter sub-image, they can be connected in sequence along the size of the position index value of each center point to obtain the center line of the corresponding connected area (i.e., the connected area corresponding to the filter image). This method has a small amount of calculation and can improve the efficiency of three-dimensional reconstruction; of course, the center points of the same filter sub-image can also be connected by fitting, such as fitting by using the least squares method or B-spline curve. Among them, the above-mentioned position index value refers to the X and Y coordinates of each center point in the filter sub-image.

[0113] The present invention also provides a device for extracting the center line of a multi-line laser light strip, comprising:

[0114] Connected region division unit: used to execute the above step S10, that is, to perform binarization processing on the original image containing multiple laser light strips to obtain a binary image; to perform connected region analysis on the binary image to obtain multiple connected regions;

[0115] Centerline extraction unit: used to execute the above step S20, that is, respectively determine the sub-image to be processed and the mask sub-image corresponding to each connected area, multiply the pixel value of the corresponding pixel in the corresponding sub-image to be processed and the mask sub-image to obtain a filtered sub-image; respectively extract the center point of the laser line segment in each of the filtered sub-images by Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segment in each filtered sub-image, that is, obtain the center position of each laser line segment in the original image;

[0116] Among them, the sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area; the mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of the pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

[0117] The specific implementation methods of the above-mentioned units can refer to the embodiments corresponding to the above-mentioned extraction method, which will not be repeated here.

[0118] It should be noted that the original image, the sub-image to be processed, the mask sub-image and the filtered sub-image involved in the present invention are usually present in the form of a matrix or array in the processing program, and these matrices or arrays have positions and elements corresponding to the positions (i.e., pixel values). Therefore, the corresponding pixels or corresponding elements described in this article are multiplied by the elements or pixel values ​​at the corresponding positions. Among them, the sub-image to be processed, the mask sub-image and the filtered sub-image are of equal size. If each is represented in the form of a matrix, the number of rows and columns of each matrix is ​​equal, and only the elements are different.

[0119] The present invention also provides a method for three-dimensional reconstruction of an object, comprising the steps of:

[0120] S100: Acquire an original image of a multi-line laser light strip projected onto the object to be measured;

[0121] S200: extracting the center position of each laser line segment in the original image using the extraction method described in any of the above embodiments;

[0122] S300: Perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of each laser line segment.

[0123] The present invention also provides a three-dimensional object reconstruction system, which can be used to implement the above-mentioned three-dimensional object reconstruction method. The three-dimensional object reconstruction system includes:

[0124] An acquisition module, used to execute the above step S100, that is, to acquire an original image of the multi-line laser light strip projected onto the object to be measured;

[0125] A centerline extraction module, comprising the above-mentioned extraction device, is used to perform the above-mentioned step S200, that is, to extract the center position of each laser line segment in the original image by using the extraction method described in any of the above-mentioned embodiments;

[0126] The three-dimensional reconstruction module is used to execute the above step S300, that is, to perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of each laser line segment.

[0127] The three-dimensional reconstruction method of the present invention can improve the accuracy and efficiency of extracting the center of the laser light strip. Therefore, when performing three-dimensional reconstruction of the object to be measured, a higher-precision model of the object to be measured can be obtained, making it closer to the actual object, and can improve the real-time performance of the three-dimensional reproduction, and can also improve the efficiency of detection when used for detection.

[0128] The present invention also provides a laser scanner, which can be a non-handheld laser scanner or a handheld laser scanner. The handheld laser scanner is a portable device used to capture the three-dimensional geometric information of the surface of an object by projecting multiple laser lines or laser beams onto the surface of the object, and then using a camera or sensor to capture the reflected light, and reconstructing the three-dimensional model of the object through image processing and calculation. Specifically, Fig.10 As shown, the laser scanner includes:

[0129] Two multi-line lasers 1 are provided for projecting multi-line laser lines to the object to be measured, that is, each multi-line laser 1 projects a multi-line laser line, and the multi-line laser lines projected by the multi-line lasers at two different positions are at a certain angle to each other;

[0130] The acquisition device 2 acquires the image of the multi-line laser light on the object to be measured to obtain the original image. The acquisition device 2 can be specifically a camera, and two cameras are provided for binocular acquisition.

[0131] The processing device 3 includes the above-mentioned object three-dimensional reconstruction system, that is, it processes the original image to perform three-dimensional reconstruction of the object to be measured.

[0132] Among them, the acquisition device 2 is signal-connected with the processing device 3 .

[0133] The multi-line laser 1 can emit multiple laser lines, which can be 5, 7, 11, 13 or 25. Two multi-line lasers 1 can be projected alternately, and then two acquisition devices 2 (such as Fig.10 The upper and lower acquisition devices 2) shown are used to synchronously acquire a multi-line laser pattern projected by a single multi-line laser.

[0134] The processing device 3 can specifically use the extraction method described in any of the above embodiments to process the original images collected by each collection device 2, obtain the center position of each laser light strip in each original image, and calculate the depth information of the object surface according to the principle of optical triangulation to perform three-dimensional reconstruction.

[0135] The laser scanner may further include a filter, which is disposed in front of the lens of the acquisition device 2 to filter out interference from non-blue light in the background, thereby improving the accuracy of subsequent extraction of the center of each light strip.

[0136] The present invention also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the method for extracting the center of a multi-line laser light strip as described in any of the above embodiments, or implements the method for three-dimensional reconstruction of an object as described above.

[0137] It should be noted that the computer-readable storage medium described in the embodiments of the present disclosure is not limited to the above-mentioned embodiments, and may also be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device.

[0138] It will be appreciated by those skilled in the art that, under the premise of no conflict, the above-mentioned preferred solutions can be freely combined and superimposed. Among them, the flowcharts and block diagrams in the accompanying drawings illustrate the possible architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a part of a code, and the module, program segment, or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings, for example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions. The numbering of each step in this article is only for the convenience of explanation and reference, and is not used to limit the order of execution. The specific execution order is determined by the technology itself, and those skilled in the art can determine various allowable and reasonable orders based on the technology itself.

[0139] It should be understood that the above-mentioned embodiments are merely illustrative and not restrictive. Without departing from the basic principles of the present invention, various obvious or equivalent modifications or substitutions that can be made by those skilled in the art to the above-mentioned details will all be included in the scope of the claims of the present invention.

Claims

1. A method for extracting the center of a multi-line laser stripe, characterized in that: Includes steps: S10: performing binarization processing on the original image containing multiple laser light strips to obtain a binarized image; performing connected region analysis on the binarized image to obtain multiple connected regions; S20: respectively determining the sub-image to be processed and the mask sub-image corresponding to each of the connected regions, and multiplying the pixel values ​​of the corresponding pixels in the sub-image to be processed and the mask sub-image to obtain a filtered sub-image; The center points of the laser line segments in each of the filter sub-images are extracted respectively by using the Gaussian weighted grayscale centroid method, so as to obtain multiple centers of the laser line segments in each of the filter sub-images, that is, to obtain the center positions of the laser line segments in the original image; in, The sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area; The mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

2. The extraction method according to claim 1, characterized in that In the step S20, before determining the sub-image to be processed and the mask sub-image corresponding to each of the connected regions, the connected regions with the number of pixels less than a preset threshold are first filtered out.

3. The extraction method according to claim 2, characterized in that The preset threshold is 10-20.

4. The extraction method according to claim 1, characterized in that In the step S20, before determining the to-be-processed sub-image and the mask sub-image corresponding to each of the connected regions, the step further includes: setting the region index value of each of the connected regions to i; wherein i is 1, 2, ..., N, and N is the number of connected regions; Determining the mask sub-image corresponding to the connected region with the region index value i includes: It is determined whether the region index value corresponding to each pixel in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area is i. If not, the pixel value of the pixel is set to 0, otherwise the pixel value of the pixel remains unchanged.

5. The extraction method according to claim 4, characterized in that In the step S20, each of the connected regions is processed in parallel.

6. The extraction method according to any one of claims 1 to 5, characterized in that In the step S20, the center points of the laser line segments in the filter sub-images are extracted respectively using the Gaussian weighted grayscale centroid method, which specifically includes the steps of: S251: determining a Gaussian weight matrix of the filter sub-image according to the following Gaussian weight function w(I(x, y)), and multiplying the Gaussian weight matrix by corresponding elements in the filter sub-image to obtain a first matrix; Where I(x,y) is the grayscale value of the pixel at the position (x,y); S252: constructing a row window matrix and a column window matrix of the same size as the filtered sub-image, and multiplying the row window matrix and the column window matrix with corresponding elements in the first matrix respectively to obtain a second matrix and a third matrix with position weight and Gaussian weight information; The element values ​​of each column in each row of the row window matrix are the same, which are the row position index values ​​of the filter sub-image corresponding to the row; the element values ​​of each row in each column of the column window matrix are the same, which are the column position index values ​​of the filter sub-image corresponding to the column; S253: sum the second matrix and the first matrix by columns, and divide the sums of corresponding columns to obtain an initial column coordinate set of the laser center processed by row index as weight; sum the third matrix and the first matrix by rows, and divide the sums of corresponding rows to obtain an initial row coordinate set of the laser center processed by column index as weight. S254: determining a preliminary center point set of the laser line segment in the filter sub-image according to the initial column coordinate set and the initial row coordinate set, wherein if the number of columns of the filter sub-image is greater than or equal to the number of rows, determining the preliminary center point set according to step S2541; if the number of rows of the filter sub-image is greater than the number of columns, determining the preliminary center point set according to step S2542; S2541: round down each element value in the initial column coordinate set to obtain a first intermediate sequence set; average the corresponding element values ​​of the initial column coordinate set and the first intermediate sequence set respectively, as the row coordinates of each center point in the initial center point set; average each element value in the first intermediate sequence set and the element value in the initial row coordinate set with the element value as the position index, as the column coordinates of each center point in the initial center point set; S2542: round down each element value in the initial row coordinate set to an integer to obtain a second intermediate sequence set, recorded as X={x1, x2, ..., xM}; average the corresponding element values ​​of the initial row coordinate set and the second intermediate sequence set respectively, as the column coordinates of each center point in the initial center point set; average each element value in the second intermediate sequence set and the element value in the initial column coordinate set with the element value as the position index, as the row coordinates of each center point in the initial center point set; S255: Judge each center point in the initial center point set. For each center point in the initial center point set, select the top, bottom, left and right four pixels of the center point coordinates on the filter sub-image, and calculate the average grayscale value of the four pixels. If the average value is greater than the preset grayscale value, use it as the confirmed center point of the laser line segment of the filter sub-image, otherwise discard the center point. Each of the confirmed center points is the multiple centers of the laser line segment of the sub-image to be processed.

7. The extraction method according to claim 1, characterized in that In the step S20, after obtaining the center of the laser line segment in each of the filter sub-images, the center points of the same filter sub-image are sequentially connected along the size of their position index values ​​to obtain the center line of the corresponding connected area.

8. The extraction method according to claim 1, characterized in that In the step S10, the original image is first subjected to Gaussian filtering or median filtering to obtain a pre-processed image, and then the pre-processed image is subjected to binarization processing; In the step S20, the sub-image to be processed corresponding to each of the connected regions is the portion of the pre-processed image corresponding to the minimum circumscribed rectangle of the connected region.

9. A device for extracting the center of a multi-line laser light strip, characterized in that: include: Connected region division unit: used for binarizing the original image containing multiple laser light strips to obtain a binary image; Performing connected region analysis on the binary image to obtain multiple connected regions; A center extraction unit: used for respectively determining the sub-image to be processed and the mask sub-image corresponding to each of the connected regions, and multiplying the pixel values ​​of the corresponding pixels in the sub-image to be processed and the mask sub-image to obtain a filtered sub-image; The center points of the laser line segments in each of the filter sub-images are extracted respectively by using the Gaussian weighted grayscale centroid method to obtain multiple centers of the laser line segments in each of the filter sub-images, that is, to obtain the center position of each laser line segment in the original image; in, The sub-image to be processed corresponding to the connected area is the part of the original image corresponding to the minimum circumscribed rectangle of the connected area; The mask sub-image corresponding to the connected area is obtained by setting the pixel values ​​of pixels outside the connected area in the part of the binary image corresponding to the minimum circumscribed rectangle of the connected area to 0, and keeping the other pixel values ​​unchanged.

10. A method for three-dimensional reconstruction of an object, characterized in that: Includes steps: S100: Acquire an original image of a multi-line laser light strip projected onto the object to be measured; S200: extracting the center position of each laser line segment in the original image using the extraction method described in any one of claims 1 to 8; S300: Perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of the laser line segments.

11. A three-dimensional object reconstruction system, characterized in that: include: An acquisition module, used to acquire an original image of the multi-line laser light strips projected onto the object to be measured; A centerline extraction module, used to extract the center position of each laser line segment in the original image by using the extraction method described in any one of claims 1 to 8; The three-dimensional reconstruction module is used to perform three-dimensional reconstruction of the object to be measured based on the obtained center positions of each of the laser line segments.

12. A laser scanner, characterized in that: include: A multi-line laser, two of which are provided, for projecting a multi-line laser line toward the object to be measured; An acquisition device is used to acquire images of multiple laser lines on the object to be measured to obtain an original image; A processing device, comprising the object three-dimensional reconstruction system according to claim 10; Wherein, the acquisition device is signal-connected to the processing device.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for extracting the center of a multi-line laser light strip as described in any one of claims 1 to 9 is implemented, or the method for three-dimensional reconstruction of an object as described in claim 10 is implemented.

Images