Intelligent hemagglutination detection method and device based on machine vision and storage medium

Abstract

The invention relates to an intelligent blood clotting detection method and device based on machine vision and a storage medium. The method comprises the following steps: acquiring image information of a blood clotting plate as a first image; performing denoising processing on the first image to obtain a second image; performing image segmentation processing, binarization and normalization processing on the second image to obtain a normalized binary image group of each hole site of the blood clotting plate as a third image group; performing similarity matching on the preprocessed images of each hole site of the third image group through five training sets formed by training through an SVM classifier algorithm in advance, and calculating to obtain the training set with the highest similarity matching degree; and marking the detection result of each hole site according to the label corresponding to the training set, and packaging and storing the image in the detection process and the corresponding detection result. According to the invention, the automation degree of the coagulometer is greatly improved, and the accuracy of a detection result is ensured. The method is applied to the field of hemagglutination detection.

Description

Technical Field

[0001] The present invention relates to the field of blood coagulation detection, and in particular to an intelligent blood coagulation detection method, device and storage medium based on machine vision. Background Art

[0002] The operation process of the coagulation detector during the experiment is roughly as follows: according to the different experimental objects, in accordance with national standards and specifications, add the specified amount of various reagents and serum, conduct the corresponding coagulation experiment, and finally interpret the image of each well of the coagulation plate to obtain the result of the experiment.

[0003] Most of the coagulation analyzers on the market today use manual or semi-automatic methods to determine the results. The main disadvantage of the manual method is the high cost, because it costs a lot to train a person who can accurately and skillfully determine the test results. In addition, the test results may fluctuate due to personal factors. The convenience of the semi-automatic method is also affected by the immature degree of automation. Summary of the invention

[0004] The purpose of the present invention is to solve at least one of the deficiencies in the prior art and to provide an intelligent blood coagulation detection method, device and storage medium based on machine vision.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] An intelligent blood coagulation detection method based on machine vision is proposed, including:

[0007] Step 110, acquiring image information of a blood coagulation plate as a first image;

[0008] Step 120: De-noising the first image to obtain a second image;

[0009] Step 130, performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group;

[0010] Step 140, performing similarity matching on the preprocessed images of each hole position of the third image group using five training sets formed by pre-training the SVM classifier algorithm, and calculating the training set with the highest similarity matching degree;

[0011] Step 150: Mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

[0012] Furthermore, the process of performing denoising on the first image to obtain the second image in the above step 120 specifically includes the following:

[0013] Step 101, extracting contour lines from the first image, and combining the extracted contour lines with an empty template image of the same size as the original image to form a fourth image;

[0014] Step 102: Perform image restoration on the first image to obtain a fifth image;

[0015] Step 103, performing image enhancement processing on the restored fifth image to obtain a sixth image;

[0016] Step 104: superimpose the first image and the sixth image to obtain a second image, and output the second image as a target image.

[0017] Furthermore, the above method of extracting the contour line in the first image is to extract the image contour feature through the Canny operator, which specifically includes the following:

[0018] Step 201, performing image smoothing processing on the first image using a Gaussian filter;

[0019] Step 202: Calculate the magnitude and direction of the gradient of the smoothed image using the finite difference of the first-order partial derivative;

[0020] Step 203, performing non-maximum suppression processing on the gradient amplitude of the smoothed image;

[0021] Step 204: perform dual threshold algorithm detection and connection edge processing on the image processed in step 203.

[0022] Furthermore, the above-mentioned method of performing image restoration on the first image to obtain the fifth image specifically includes the following:

[0023] Step 301, P represents all parts of the first image, Q is determined to be the part of the first image that needs to be repaired, and PQ is the target part that does not need to be repaired;

[0024] Step 302: Calculate each pixel point a on the edge of the portion Q to be repaired of the first image, and calculate the block M to be repaired centered at the pixel point a. a Repair Factor A:

[0025] A(a)=C(a)D(a),

[0026] Where C(a) is the block M to be repaired a The confidence factor term, D(a) is the block to be repaired M a The data factor term is Where |M a | is the block to be repaired Ma The area of ​​​​the , α is the normalization factor; represents the unit vector at point a that is perpendicular to the edge of the part Q to be repaired;

[0027] Step 303: Find the pixel point corresponding to the MAX A(a) value among the edge points of the portion to be repaired Q, and first repair the block to be repaired centered on the pixel point corresponding to the MAX A(a) value. a The repair factor is set to 0;

[0028] Step 304: Repeat step 303 until the part Q to be repaired is completely repaired.

[0029] Furthermore, the repair process in step 303 includes the following:

[0030] Step 401: Calculate the block M to be repaired a The Euclidean distance d between the grayscale value of the known pixel in the image and the grayscale value of the corresponding block in the part of the image that does not need to be repaired, and the block with the minimum Euclidean distance is taken as the block to be repaired M a The N best matching image blocks a ;

[0031] Step 402: According to the best matching image block N a The gray value of the pixel in the repair block M a The gray value of the pixel point is filled in, and after filling, the block to be repaired M a Mark the part that does not need repair;

[0032] Step 403: reprocess the image processed in step 402 according to steps 401 to 402 until the part to be repaired is completely repaired.

[0033] Furthermore, the formation of the five training sets in the above step 140 specifically includes the following:

[0034] First, determine the training image acquisition criteria corresponding to the five training sets.

[0035] The hole image characteristics corresponding to the first training set are that the red blood cells are agglutinated in sheets and evenly spread over the bottom of the hole, with irregularities or wrinkles on the edges, and the label corresponding to the training set is ++++.

[0036] The hole image characteristics of the second training set are that the red blood cells are agglutinated in sheets, and the agglutination area is smaller than that of the first training set. The label corresponding to the second training set is +++.

[0037] The hole position image characteristics corresponding to the third training set are that the red blood cells are agglutinated in sheets, and some of the deposits are in the shape of dots in the middle of the bottom of the hole. The label corresponding to the third training set is ++.

[0038] The characteristics of the well image corresponding to the fourth training set are that most of the red blood cells sink to the center of the well bottom, with a few scattered agglutinated particles around them, and the label corresponding to the fourth training set is +.

[0039] The hole image characteristics corresponding to the fifth training set are that the red blood cells form dots at the bottom of the hole with neat edges, and the label corresponding to the fifth training set is -.

[0040] The training process is as follows,

[0041] Step 501: collect hole position images with corresponding hole position image characteristics to prepare training sets, each training set collects no less than 100 images, and obtains images corresponding to 5 training sets according to the above acquisition criteria;

[0042] Step 502: train each image in the training set, and automatically generate a training model .xml file from the trained model;

[0043] When identifying,

[0044] Step 503: read the training model .xml file, and identify the preprocessed image of each hole position of the third image group according to the trained model to obtain a corresponding recognition result.

[0045] The present invention also proposes an intelligent blood coagulation detection device based on machine vision, the device comprising:

[0046] An initial image acquisition module, used to acquire image information of a blood coagulation plate as a first image;

[0047] A denoising module, used for performing denoising processing on the first image to obtain a second image;

[0048] a well position image acquisition module, used for performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group;

[0049] A training set matching module, used to perform similarity matching on the preprocessed images of each hole position of the third image group using 5 training sets formed by pre-training the SVM classifier algorithm, and calculate the training set with the highest similarity matching degree;

[0050] The result processing module is used to mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

[0051] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the method as described in any one of claims 1 to 6 are implemented.

[0052] The beneficial effects of the present invention are:

[0053] The present invention obtains the overall image of the coagulation plate, obtains the detection image of each well position on the coagulation plate by image segmentation and normalization, and obtains the closest training set by similarity calculation based on the trained SVM training set, and finally marks the experimental results of the corresponding well positions according to the pre-set labels of the corresponding training set, and finally retains the results for staff to view, which greatly improves the automation level of the coagulation analyzer and ensures the accuracy of the detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0055] Figure 1 Shown is a flow chart of an intelligent blood coagulation detection method based on machine vision;

[0056] Figure 2 Shown is a structural diagram of an intelligent blood coagulation detection device based on machine vision;

[0057] Figure 3 Shown are standard images corresponding to two situations of the first training set of the present invention;

[0058] Figure 4 Shown is the standard image corresponding to the second training set of the present invention;

[0059] Figure 5 Shown is the standard image corresponding to the third training set of the present invention;

[0060] Figure 6 Shown is the standard image corresponding to the fourth training set of the present invention;

[0061] Figure 7 Shown is the standard image corresponding to the fifth training set of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the embodiments and drawings to clearly and completely describe the concept, specific structure and technical effects of the present invention, so as to fully understand the purpose, scheme and effect of the present invention. It should be noted that the embodiments in this application and the features in the embodiments can be combined with each other without conflict. The same reference numerals used throughout the drawings indicate the same or similar parts.

[0063] Combination Figure 1 Embodiment 1: The present invention proposes an intelligent blood coagulation detection method based on machine vision, comprising:

[0064] Step 110, acquiring image information of a blood coagulation plate as a first image;

[0065] Step 120: De-noising the first image to obtain a second image;

[0066] Step 130, performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group;

[0067] Step 140, performing similarity matching on the preprocessed images of each hole position of the third image group using five training sets formed by pre-training the SVM classifier algorithm, and calculating the training set with the highest similarity matching degree;

[0068] Step 150: Mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

[0069] In Example 1, the whole image of the coagulation plate is obtained, and the detection image of each well on the coagulation plate is obtained by image segmentation and normalization. The similarity is calculated based on the trained SVM training set to obtain the closest training set. Finally, the experimental results of the corresponding wells are marked according to the pre-set labels of the corresponding training set, and the results are finally retained for the staff to review, which greatly improves the automation level of the coagulation analyzer and ensures the accuracy of the test results.

[0070] As a preferred implementation of this solution, the process of performing denoising on the first image to obtain the second image in the above step 120 specifically includes the following:

[0071] Step 101, extracting contour lines from the first image, and combining the extracted contour lines with an empty template image of the same size as the original image to form a fourth image;

[0072] Step 102: Perform image restoration on the first image to obtain a fifth image;

[0073] Step 103, performing image enhancement processing on the restored fifth image to obtain a sixth image;

[0074] Step 104: superimpose the first image and the sixth image to obtain a second image, and output the second image as a target image.

[0075] As a preferred implementation of this solution, the above-mentioned method of extracting the contour line in the first image is to extract the image contour feature through the Canny operator, which specifically includes the following:

[0076] Step 201, performing image smoothing processing on the first image using a Gaussian filter;

[0077] Step 202: Calculate the magnitude and direction of the gradient of the smoothed image using the finite difference of the first-order partial derivative;

[0078] Step 203, performing non-maximum suppression processing on the gradient amplitude of the smoothed image;

[0079] Step 204: perform dual threshold algorithm detection and connection edge processing on the image processed in step 203.

[0080] As a preferred implementation of this solution, the above-mentioned method of performing image restoration on the first image to obtain the fifth image specifically includes the following:

[0081] Step 301, P represents all parts of the first image, Q is determined to be the part of the first image that needs to be repaired, and PQ is the target part that does not need to be repaired;

[0082] Step 302: Calculate each pixel point a on the edge of the portion Q to be repaired of the first image, and calculate the block M to be repaired centered at the pixel point a. a Repair Factor A:

[0083] A(a)=C(a)D(a),

[0084] Where C(a) is the block M to be repaired a The confidence factor term, D(a) is the block to be repaired M a The data factor term is Where |M a | is the block to be repaired M a The area of ​​​​the , α is the normalization factor; represents the unit vector at point a that is perpendicular to the edge of the part Q to be repaired;

[0085] Step 303: Find the pixel point corresponding to the MAX A(a) value among the edge points of the portion to be repaired Q, and first repair the block to be repaired centered on the pixel point corresponding to the MAX A(a) value. aThe repair factor is set to 0;

[0086] Step 304: Repeat step 303 until the part Q to be repaired is completely repaired.

[0087] As a preferred implementation of this solution, the repair process in step 303 includes the following:

[0088] Step 401: Calculate the block M to be repaired a The Euclidean distance d between the grayscale value of the known pixel in the image and the grayscale value of the corresponding block in the part of the image that does not need to be repaired, and the block with the minimum Euclidean distance is taken as the block to be repaired M a The N best matching image blocks a ;

[0089] Step 402: According to the best matching image block N a The gray value of the pixel in the repair block M a The gray value of the pixel point is filled in, and after filling, the block to be repaired M a Mark the part that does not need repair;

[0090] Step 403: reprocess the image processed in step 402 according to steps 401 to 402 until the part to be repaired is completely repaired.

[0091] As a preferred implementation of this solution, the formation of the five training sets in the above step 140 specifically includes the following:

[0092] First, determine the training image acquisition criteria corresponding to the five training sets.

[0093] The hole image characteristics corresponding to the first training set are that the red blood cells are agglutinated in sheets and evenly spread over the bottom of the hole, with irregularities or wrinkles on the edges, and the label corresponding to the training set is ++++.

[0094] The hole image characteristics of the second training set are that the red blood cells are agglutinated in sheets, and the agglutination area is smaller than that of the first training set. The label corresponding to the second training set is +++.

[0095] The hole position image characteristics corresponding to the third training set are that the red blood cells are agglutinated in sheets, and some of the deposits are in the shape of dots in the middle of the bottom of the hole. The label corresponding to the third training set is ++.

[0096] The characteristics of the well image corresponding to the fourth training set are that most of the red blood cells sink to the center of the well bottom, with a few scattered agglutinated particles around them, and the label corresponding to the fourth training set is +.

[0097] The hole image characteristics corresponding to the fifth training set are that the red blood cells form dots at the bottom of the hole with neat edges, and the label corresponding to the fifth training set is -.

[0098] The training process is as follows,

[0099] Step 501: collect hole position images with corresponding hole position image characteristics to prepare training sets, each training set collects no less than 100 images, and obtains images corresponding to 5 training sets according to the above acquisition criteria;

[0100] Step 502: train each image in the training set, and automatically generate a training model .xml file from the trained model;

[0101] When identifying,

[0102] Step 503: read the training model .xml file, and identify the preprocessed image of each hole position of the third image group according to the trained model to obtain a corresponding recognition result.

[0103] Combination Figure 2 , Embodiment 2, the present invention also proposes an intelligent blood coagulation detection device based on machine vision, the device comprising:

[0104] An initial image acquisition module, used to acquire image information of a blood coagulation plate as a first image;

[0105] A denoising module, used for performing denoising processing on the first image to obtain a second image;

[0106] a well position image acquisition module, used for performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group;

[0107] A training set matching module, used to perform similarity matching on the preprocessed images of each hole position of the third image group using 5 training sets formed by pre-training the SVM classifier algorithm, and calculate the training set with the highest similarity matching degree;

[0108] The result processing module is used to mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

[0109] In Example 2, by acquiring the overall image of the coagulation plate, the detection image of each well on the coagulation plate is obtained by image segmentation and normalization processing, and the similarity calculation is performed based on the trained SVM training set to obtain the closest training set, and finally the experimental results of the corresponding wells are marked according to the pre-set labels of the corresponding training set, and finally the results are retained for the staff to review, which greatly improves the automation level of the coagulation analyzer and ensures the accuracy of the test results.

[0110] The present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and is characterized in that when the computer program is executed by a processor, the steps of the method as described in any one of claims 1 to 6 are implemented.

[0111] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one place or distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0112] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of software functional modules.

[0113] If the integrated module is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0114] Although the description of the present invention has been quite detailed and specifically describes several described embodiments, it is not intended to be limited to any of these details or embodiments or any particular embodiment, but should be regarded as providing a broad possible interpretation of these claims in view of the prior art by reference to the appended claims, thereby effectively covering the intended scope of the present invention. In addition, the above description of the present invention is based on the embodiments foreseeable by the inventor, and its purpose is to provide a useful description, and those non-substantial changes to the present invention that have not yet been foreseen may still represent equivalent changes to the present invention.

[0115] The above is only a preferred embodiment of the present invention. The present invention is not limited to the above implementation. As long as the technical effect of the present invention is achieved by the same means, it should belong to the protection scope of the present invention. Within the protection scope of the present invention, its technical scheme and / or implementation method can have various modifications and changes.

Claims

1. An intelligent blood coagulation detection method based on machine vision, characterized in that: include: Step 110, acquiring image information of a blood coagulation plate as a first image; Step 120: De-noising the first image to obtain a second image; Step 130, performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group; Step 140, performing similarity matching on the preprocessed images of each hole position of the third image group using five training sets formed by pre-training the SVM classifier algorithm, and calculating the training set with the highest similarity matching degree; Step 150: Mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

2. The intelligent blood coagulation detection method based on machine vision according to claim 1, characterized in that: The process of performing denoising on the first image to obtain the second image in step 120 specifically includes the following: Step 101, extracting contour lines from the first image, and combining the extracted contour lines with an empty template image of the same size as the original image to form a fourth image; Step 102: Perform image restoration on the first image to obtain a fifth image; Step 103, performing image enhancement processing on the restored fifth image to obtain a sixth image; Step 104: superimpose the first image and the sixth image to obtain a second image, and output the second image as a target image.

3. The intelligent blood coagulation detection method based on machine vision according to claim 2, characterized in that: The above method of extracting the contour line in the first image is to extract the image contour feature through the Canny operator, which specifically includes the following: Step 201, performing image smoothing processing on the first image using a Gaussian filter; Step 202: Calculate the magnitude and direction of the gradient of the smoothed image using the finite difference of the first-order partial derivative; Step 203, performing non-maximum suppression processing on the gradient amplitude of the smoothed image; Step 204: perform dual threshold algorithm detection and connection edge processing on the image processed in step 203.

4. The intelligent blood coagulation detection method based on machine vision according to claim 3, characterized in that: The above-mentioned method of performing image restoration on the first image to obtain the fifth image specifically includes: the following: Step 301, P represents all parts of the first image, Q is determined to be the part of the first image that needs to be repaired, and PQ is the target part that does not need to be repaired; Step 302: Calculate each pixel point a on the edge of the portion Q to be repaired of the first image, and calculate the block M to be repaired centered at the pixel point a. a Repair Factor A: A(a)=C(a)D(a), Where C(a) is the block M to be repaired a The confidence factor term, D(a) is the block to be repaired M a The data factor term is Where |M a | is the block to be repaired M a The area of ​​​​the , α is the normalization factor; represents the unit vector at point a that is perpendicular to the edge of the part Q to be repaired; Step 303: Find the pixel point corresponding to the MAX A(a) value among the edge points of the portion to be repaired Q, and first repair the block to be repaired centered on the pixel point corresponding to the MAX A(a) value. a The repair factor is set to 0; Step 304: Repeat step 303 until the part Q to be repaired is completely repaired.

5. The intelligent blood coagulation detection method based on machine vision according to claim 4, characterized in that: The repair process in step 303 includes: the following: Step 401: Calculate the block M to be repaired a The Euclidean distance d between the grayscale value of the known pixel in the image and the grayscale value of the corresponding block in the part of the image that does not need to be repaired, and the block with the minimum Euclidean distance is taken as the block to be repaired M a The N best matching image blocks a ; Step 402: According to the best matching image block N a The gray value of the pixel in the repair block M a The gray value of the pixel point is filled in, and after filling, the block to be repaired M a Mark the part that does not need repair; Step 403: reprocess the image processed in step 402 according to steps 401 to 402 until the part to be repaired is completely repaired.

6. The intelligent blood coagulation detection method based on machine vision according to claim 1, characterized in that: The formation of the five training sets in the above step 140 specifically includes the following: First, determine the training image acquisition criteria corresponding to the five training sets. The hole image characteristics corresponding to the first training set are that the red blood cells are agglutinated in sheets and evenly spread over the bottom of the hole, with irregularities or wrinkles on the edges, and the label corresponding to the training set is ++++. The hole image characteristics of the second training set are that the red blood cells are agglutinated in sheets, and the agglutination area is smaller than that of the first training set. The label corresponding to the second training set is +++. The hole position image characteristics corresponding to the third training set are that the red blood cells are agglutinated in sheets, and some of the deposits are in the shape of dots in the middle of the bottom of the hole. The label corresponding to the third training set is ++. The characteristics of the well image corresponding to the fourth training set are that most of the red blood cells sink to the center of the well bottom, with a few scattered agglutinated particles around them, and the label corresponding to the fourth training set is +. The hole image characteristics corresponding to the fifth training set are that the red blood cells form dots at the bottom of the hole with neat edges, and the label corresponding to the fifth training set is -. The training process is as follows, Step 501: collect hole position images with corresponding hole position image characteristics to prepare training sets, each training set collects no less than 100 images, and obtains images corresponding to 5 training sets according to the above acquisition criteria; Step 502: train each image in the training set, and automatically generate a training model .xml file from the trained model; When identifying, Step 503: read the training model .xml file, and identify the preprocessed image of each hole position of the third image group according to the trained model to obtain a corresponding recognition result.

7. An intelligent blood coagulation detection device based on machine vision, characterized in that: The device comprises: An initial image acquisition module, used to acquire image information of a blood coagulation plate as a first image; A denoising module, used for performing denoising processing on the first image to obtain a second image; a well position image acquisition module, used for performing image segmentation, binarization and normalization processing on the second image to obtain a normalized binary image group of each well position of the blood coagulation plate as a third image group; A training set matching module, used to perform similarity matching on the preprocessed images of each hole position of the third image group using 5 training sets formed by pre-training the SVM classifier algorithm, and calculate the training set with the highest similarity matching degree; The result processing module is used to mark the detection result of each hole position according to the label corresponding to the training set, and package and save the image in the detection process and the corresponding detection result.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images