Tips location information recognition method and terminal based on camera

Abstract

The present invention relates to a camera-based tip location information recognition method and terminal. The method comprises: acquiring an initial image of a jig plate; converting the initial image into a two-tone image; dividing the two-tone image into regions according to the distribution of holes in the jig plate; and determining whether a hole in each region carries a tip based on the number of preset color pixels within each region, obtaining and outputting a judgment result, where the preset color pixels are pixels containing one of the two-tone colors. The present invention enables automated recognition of tip location information, thereby avoiding the problems caused by manual recognition.

Description

Technical Field

[0001] The present invention belongs to the technical field of machine vision detection, and in particular relates to a camera-based Tips location information recognition method and terminal. Background Art

[0002] Existing jigs have several holes, each of which may hold a tip. Before injecting reagents or samples into the tips, it's important to first determine which positions on the jig are loaded with tips and which are not. Without this tip check, reagents or samples may be injected into empty holes, potentially causing biochemical contamination in the operating environment.

[0003] The existing technology relies on manual determination of whether the equipment is fully loaded. If it is empty, the empty position needs to be manually specified. This not only increases the workload of the operator, but also may cause human misjudgment and cause biochemical contamination problems of the equipment. Summary of the Invention

[0004] The purpose of the present invention is to provide a camera-based Tips location information recognition method and terminal to achieve automatic recognition of Tips location information, thereby avoiding the problems caused by manual recognition.

[0005] In order to solve the above technical problems, the present invention discloses a camera-based Tips location information recognition method, comprising:

[0006] Get the initial image of the jig plate;

[0007] converting the initial image into a two-tone image;

[0008] Dividing the two-tone image into regions according to the hole distribution of the jig plate;

[0009] Whether the hole in each area carries Tips is determined based on the number of preset color pixels in each area, and a determination result is obtained and output. The preset color pixels are pixels containing one of the two-tone colors.

[0010] Furthermore, the obtaining of the initial image of the jig plate includes:

[0011] intercepting position information of the jig plate according to the color difference between the jig plate in an empty state and the shooting environment;

[0012] An initial image is captured based on the position information of the jig plate.

[0013] Furthermore, intercepting the position information of the jig plate according to the color difference between the jig plate in an empty state and the shooting environment includes:

[0014] Acquire a debugging image of the fixture plate in an unloaded state, wherein the debugging image is an image of the fixture plate in an unloaded state captured in an ambient color different from the color of the fixture plate;

[0015] Performing color clustering on the debugging image, where the number of clusters is 2 and the initial color values ​​are the color of the fixture board in an unloaded state and the ambient color of the shooting environment, respectively;

[0016] The coordinate extreme values ​​of all pixels clustered as the color of the jig plate are extracted to obtain the position information of the jig plate.

[0017] Furthermore, converting the initial image into a two-tone image comprises:

[0018] Gray-scale the initial image to obtain a grayscale image;

[0019] The grayscale image is processed according to two preset colors to obtain a two-tone image.

[0020] Furthermore, the processing of the grayscale image according to two preset colors to obtain a two-tone image includes:

[0021] Obtaining an average grayscale value of the grayscale image, and using the average grayscale value as a grayscale threshold;

[0022] The color of each pixel on the grayscale image is converted into two preset colors according to whether its grayscale value is greater than a grayscale threshold, thereby obtaining a two-tone image.

[0023] Furthermore, before dividing the two-tone image into regions according to the hole distribution of the jig plate, the method further includes:

[0024] An opening operation is performed on the two-tone image, where the morphological structure element of the opening operation is the cross-sectional shape of Tips.

[0025] Furthermore, dividing the two-tone image into regions according to the hole distribution of the jig plate includes:

[0026] The two-tone image is evenly divided into rectangular areas having the same number of holes according to the hole distribution of the jig plate.

[0027] Furthermore, judging whether the hole in each area carries a Tip according to the number of pixels of a preset color in the area includes:

[0028] Whether the hole in the area carries a Tip is determined based on whether the number of preset color pixels in each area is greater than a number threshold, and the number threshold is set by a percentage of the number of pixels in the area.

[0029] Furthermore, the two-tone image is a black and white image.

[0030] In order to solve the above technical problems, the present invention also discloses a camera-based Tips location information recognition terminal, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the above-mentioned camera-based Tips location information recognition method is implemented.

[0031] This invention obtains an initial image of the jig plate, converts the initial image into a two-tone image, divides the two-tone image into regions according to the distribution of holes in the jig plate, and finally determines whether the holes in the region contain tips based on the number of preset color pixels in each region, thereby realizing automatic recognition of tip location information and avoiding problems caused by manual recognition. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 Schematic diagram of the process of the camera-based Tips location information recognition method according to an embodiment of the present invention;

[0033] Figure 2 is a schematic diagram of a grayscale image involved in an embodiment of the present invention;

[0034] Figure 3 is a schematic diagram of a black and white image involved in an embodiment of the present invention;

[0035] Figure 4 A schematic diagram of a judgment result in a black and white mode according to an embodiment of the present invention;

[0036] Figure 5 2 is a schematic structural diagram of a camera-based Tips location information recognition terminal according to an embodiment of the present invention.

[0037] Description of labels:

[0038] 1. Camera-based Tips location information recognition terminal; 2. Processor; 3. Memory. DETAILED DESCRIPTION

[0039] The present invention is further described in detail below through examples so that those skilled in the art can implement the invention with reference to the description.

[0040] It should be understood that terms such as “having”, “including” and “comprising” used herein do not preclude the existence or addition of one or more other elements or combinations thereof.

[0041] The specific embodiment of the present invention can be applied when it is necessary to perform operations such as injecting reagents or samples on the tips on the fixture plate. The present invention can help accurately obtain the position information of the tips on the fixture plate.

[0042] Example 1

[0043] In this embodiment, in order to capture the image of the fixture plate, a camera needs to be installed on the original device. Therefore, the description of the installed camera is as follows:

[0044] (1) Camera installation position: The camera is directly installed on the reagent injection arm. Therefore, there is no need to add new fixing parts to fix the camera to avoid interfering with the movement of the reagent injection arm.

[0045] (2) Camera shooting direction: directly below or facing the fixture plate.

[0046] (3) Camera shooting height: 10cm~15cm, ensuring that the entire fixture plate is in the field of view and the edge of the fixture plate has no significant deformation.

[0047] (4) Photo shooting timing: Before the first injection of the jig plate to be injected, when the injection arm moves to the top of the jig plate, pause for a moment and take the photo.

[0048] (5) Shooting environment: Ordinary white light, but the background color should not be white and should be significantly different from the color of the fixture board.

[0049] like Figures 2 to 4 It can be seen that the fixture plate in this embodiment is a 96-well plate with 96 holes thereon. In other embodiments, corresponding fixture plates can be placed according to user needs.

[0050] Therefore, combined Figures 1 to 4 This embodiment provides a camera-based Tips location information recognition method, including:

[0051] Step S1, obtaining an initial image of the jig plate;

[0052] In this embodiment, step S1 includes:

[0053] Step S11, intercepting the position information of the jig plate according to the color difference between the jig plate in an empty state and the shooting environment;

[0054] Wherein, step S11 includes:

[0055] Step S111: Acquire a debugging image of the fixture plate in an empty state, where the debugging image is an image of the fixture plate in an empty state captured in an ambient color different from the color of the fixture plate;

[0056] Step S112: Perform color clustering on the debugging image, where the number of clusters is 2 and the initial color values ​​are the color of the fixture plate in an unloaded state and the ambient color of the shooting environment, respectively;

[0057] Step S113 : extracting the coordinate extreme values ​​of all pixels clustered as the color of the jig plate to obtain the position information of the jig plate.

[0058] The extreme values ​​of coordinates include (minX, minY), (minX, maxY), (maxX, minY), and (maxX, maxY), so that the captured initial image is a rectangular image.

[0059] Among them, when batch producing equipment, there will inevitably be some errors in the production and debugging process, making each device slightly different. Step S11 should be performed before leaving the factory to reduce the workload of the production company's technical staff.

[0060] Step S12: capturing an initial image based on the position information of the jig plate.

[0061] Since the relative position relationship between the camera and the object on the same device is fixed and the size of the object is fixed during each detection, the cropped vertex position is also fixed. The rectangular image can be captured with fixed vertex coordinates during the debugging process. That is, after the vertex coordinates of the 96-well plate are obtained in step S11, it is delivered to the user for use, so that the user only needs to execute step S12 during use after leaving the factory, thereby avoiding failures caused by unpredictable user usage environment.

[0062] It should be noted that, in this embodiment, the 96-well plate is blue or dark green in the initial image.

[0063] Step S2, converting the initial image into a two-tone image;

[0064] In this embodiment, step S2 includes:

[0065] Step S21, grayscale processing is performed on the initial image to obtain a grayscale image;

[0066] It should be noted that, in this embodiment, the 96-well plate appears blue or dark green in the initial image.

[0067] Substitute the obtained initial image into the following formula: Gray = 0.30R + 0.59G + 0.11B, where Gray is the calculated gray value, and R, G, and B are the values ​​of the red, green, and blue channels of the color image. Then we get Figure 2 In other embodiments, other grayscale formulas may be used for grayscale processing of the initial image.

[0068] Step S22: Process the grayscale image according to two preset colors to obtain a two-tone image.

[0069] In this embodiment, the two-tone image is a black and white image. In other embodiments, the two-tone image can be distinguished by two colors selected by the user.

[0070] Therefore, step S22 is to convert the grayscale image into a black and white image, which includes:

[0071] Step S221, obtaining the average grayscale value of the grayscale image, and using the average grayscale value as the grayscale threshold;

[0072] like Figure 2 The grayscale threshold shown is 0.511.

[0073] Step S222 : converting the color of each pixel on the grayscale image into two preset colors according to whether its grayscale value is greater than a grayscale threshold, thereby obtaining a two-tone image.

[0074] That is, if the grayscale value of each pixel on the grayscale image is greater than 0.511, it will be converted to black, and if it is less than 0.511, it will be converted to white, forming a black and white image as shown in Figure 3.

[0075] Step S3: performing an opening operation on the two-tone image, where the morphological structure element of the opening operation is the cross-sectional shape of Tips.

[0076] This involves first eroding and then dilating a black and white image, smoothing the image's contours, reducing narrow areas, and removing thin protrusions. In this example, the cross-sectional shape of the tips is circular, so the morphological structuring element of the opening operation is a circle. The radius of the circle depends on the image size; a reasonable value is 0.1% of the width. In this example, the image has a horizontal width of 2640 pixels, so the radius is 3 pixels.

[0077] Step S4: Divide the two-tone image into regions according to the hole distribution of the jig plate;

[0078] In this embodiment, step S4 includes:

[0079] The two-tone image is divided into rectangular areas with the same number of holes according to the hole distribution of the fixture plate.

[0080] That is, the entire image is evenly divided into 8 rows and 12 columns, a total of 96 areas.

[0081] Step S5: Determine whether the hole in each area carries a Tip based on the number of preset color pixels in each area, and obtain and output a determination result. The preset color pixels are pixels containing one of the two-tone colors.

[0082] In this embodiment, determining whether a hole in each area carries a tip based on the number of pixels of a preset color in each area includes:

[0083] Whether the hole in the area carries a tip is determined by whether the number of preset color pixels in each area is greater than a threshold. The threshold is set as a percentage of the number of pixels in the area.

[0084] In this embodiment, the default color pixel is white. That is, the number of white pixels in each region is counted. If the number is greater than a threshold, the location is considered to have a tip; otherwise, the location is considered unloaded. The threshold is within a wide range, such as 3%-10% of the total number of pixels in the region. In this example, the image is 2460×1760 pixels, divided into 96 regions. Each region has 48,400 pixels, and the threshold is 3000, which is approximately 6.2% of the total number of pixels in the region.

[0085] The final judgment result is referenced Figure 4 shown.

[0086] But it should be noted that, because Figure 4 It is a black and white image that meets the patent requirements, and therefore cannot display a true color image. In the actual color image, the area borders for tips are green, and the unloaded areas are actually distinguished by rectangular red borders. That is, each area has an area border, and different colors are given to distinguish it according to whether it carries tips.

[0087] Similarly, after obtaining the judgment results on the black and white image, this embodiment will eventually generate the judgment results on the initial image, that is, each hole in the initial image will generate an area edge line according to the divided area, and will be given different colors to distinguish it according to whether it carries Tips.

[0088] In this way, the automatic recognition of Tips location information is achieved, thereby avoiding the problems caused by manual recognition, which reduces the workload of operators and avoids biochemical contamination of equipment caused by human misjudgment.

[0089] Example 2

[0090] like Figure 5 As shown, the camera-based Tips location information recognition terminal 1 includes a memory 3, a processor 2, and a computer program stored in the memory 3 and executable on the processor 2. When the processor 2 executes the computer program, the camera-based Tips location information recognition method in the first embodiment is implemented.

[0091] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the description and implementation methods. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. The camera-based Tips location information recognition method is characterized by: include: Get the initial image of the jig plate; converting the initial image into a two-tone image; Dividing the two-tone image into regions according to the hole distribution of the jig plate; Determine whether the hole in each area carries a Tip based on the number of preset color pixels in each area, and obtain and output a determination result, wherein the preset color pixels are pixels containing one of the two-tone colors; The step of dividing the two-tone image into regions according to the hole distribution of the jig plate includes: Dividing the two-tone image into rectangular areas having the same number of holes according to the hole distribution of the jig plate; The obtaining of the initial image of the fixture plate includes: intercepting position information of the jig plate according to the color difference between the jig plate in an empty state and the shooting environment; An initial image is captured based on the position information of the jig plate.

2. The camera-based Tips location information recognition method according to claim 1, characterized in that: The method of intercepting the position information of the jig plate according to the color difference between the jig plate in an empty state and the shooting environment includes: Acquire a debugging image of the fixture plate in an unloaded state, wherein the debugging image is an image of the fixture plate in an unloaded state captured in an ambient color different from the color of the fixture plate; Performing color clustering on the debugging image, where the number of clusters is 2 and the initial color values ​​are the color of the fixture board in an unloaded state and the ambient color of the shooting environment, respectively; The coordinate extreme values ​​of all pixels clustered as the color of the jig plate are extracted to obtain the position information of the jig plate.

3. The camera-based Tips location information recognition method according to claim 1, characterized in that: The converting the initial image into a two-tone image comprises: Gray-scale the initial image to obtain a grayscale image; The grayscale image is processed according to two preset colors to obtain a two-tone image.

4. The camera-based Tips location information recognition method according to claim 3, characterized in that: Processing the grayscale image according to two preset colors to obtain a two-tone image includes: Obtaining an average grayscale value of the grayscale image, and using the average grayscale value as a grayscale threshold; The color of each pixel on the grayscale image is converted into two preset colors according to whether its grayscale value is greater than a grayscale threshold, thereby obtaining a two-tone image.

5. The camera-based Tips location information recognition method according to claim 1, characterized in that: Before dividing the two-tone image into regions according to the hole distribution of the jig plate, the method further includes: An opening operation is performed on the two-tone image, where the morphological structure element of the opening operation is the cross-sectional shape of Tips.

6. The camera-based Tips location information recognition method according to claim 1, characterized in that: The step of determining whether a hole in each area carries a Tip according to the number of pixels of a preset color in each area includes: Whether the hole in the area carries a Tip is determined based on whether the number of preset color pixels in each area is greater than a number threshold, and the number threshold is set by a percentage of the number of pixels in the area.

7. The camera-based Tips location information recognition method according to any one of claims 1 to 6, characterized in that: The two-tone image is a black and white image.

8. A camera-based Tips location information recognition terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the camera-based Tips location information recognition method according to any one of claims 1 to 6 is implemented.

Images