WO2017223159A1 - Chemical test card automatic reading method and system - Google Patents

Chemical test card automatic reading method and system Download PDF

Info

Publication number
WO2017223159A1
WO2017223159A1 PCT/US2017/038466 US2017038466W WO2017223159A1 WO 2017223159 A1 WO2017223159 A1 WO 2017223159A1 US 2017038466 W US2017038466 W US 2017038466W WO 2017223159 A1 WO2017223159 A1 WO 2017223159A1
Authority
WO
WIPO (PCT)
Prior art keywords
image
test card
chemical test
image data
chemical
Prior art date
Application number
PCT/US2017/038466
Other languages
French (fr)
Inventor
Ling Chen
Xinhua FENG
Jie Jin
Huaiping Rong
Original Assignee
3M Innovative Properties Company
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 3M Innovative Properties Company filed Critical 3M Innovative Properties Company
Publication of WO2017223159A1 publication Critical patent/WO2017223159A1/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/44Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
    • AHUMAN NECESSITIES
    • A61MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61LMETHODS OR APPARATUS FOR STERILISING MATERIALS OR OBJECTS IN GENERAL; DISINFECTION, STERILISATION OR DEODORISATION OF AIR; CHEMICAL ASPECTS OF BANDAGES, DRESSINGS, ABSORBENT PADS OR SURGICAL ARTICLES; MATERIALS FOR BANDAGES, DRESSINGS, ABSORBENT PADS OR SURGICAL ARTICLES
    • A61L2/00Methods or apparatus for disinfecting or sterilising materials or objects other than foodstuffs or contact lenses; Accessories therefor
    • A61L2/26Accessories or devices or components used for biocidal treatment
    • A61L2/28Devices for testing the effectiveness or completeness of sterilisation, e.g. indicators which change colour
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/20Image preprocessing
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00Arrangements for image or video recognition or understanding
    • G06V10/40Extraction of image or video features
    • G06V10/50Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis

Definitions

  • the present disclosure relates to the technical field of image recognition, and in particular, to a chemical test card automatic reading method and system.
  • a high-pressure steam sterilization device is generally used for disinfecting and sterilizing medical instruments.
  • the process of disinfection if air in the sterilization device is not completely discharged or air leakage occurs, some air may reside in the sterilization device. The residual air may affect permeation of high-pressure steam to a to-be-disinfected article in the sterilization device in a steam exposure stage of the process of sterilization, thus reducing the sterilization effect.
  • a BD (Bowie-Dick) test is named after last names of inventors thereof, i.e., J. H. Bowie and J. Dick.
  • the BD test is used to evaluate the vacuum state of a sterilization device, thereby judging whether the sterilization device works normally.
  • a BD test card in a BD test pack is placed into a sterilization device together with a to-be-disinfected article, to test the sterilization device.
  • FIG. 1 shows a common BD test card
  • FIG. 2A to FIG. 2C show examples of a BD test card after high-pressure steam sterilization and disinfection.
  • a stripe region of the BD test card mainly includes two parts, namely, a reactant region and a blank region.
  • a change generated on the BD test card after being heated by a high temperature is generally the color change of the reactant region.
  • the reactant on the BD test card is originally in a light yellow color, and the reactant may change from light yellow to uniform dark brown and to black after being heated by a high temperature, while the color of the blank region is always white. If the color of the reactant is changed uniformly, as shown in FIG. 2A, it indicates that the BD test is qualified, and the sterilization device can work normally. If the color of the reactant of the BD test card is changed non-uniformly, as shown in FIG. 2B, it indicates that the BD test is unqualified, and the sterilization device has an air leakage phenomenon and cannot work normally.
  • an operator judges whether the color change effect of a BD test card reaches a designated standard with the naked eye, and such a judging method is very subjective. Different operators have different sensitivities on changes in color and brightness, and this may result in different judgment results of different operators for the same BD test card. Particularly, for an example in a critical state as shown in FIG. 2C, different operators may obtain different judgment results. Moreover, the same operator may also obtain different judgment results for the same BD test card in different time. This causes unreliability of judging the color change effect of the BD test card by naked eyes, thus bringing about potential risks for medical and health care quality management.
  • PCD Process Challenge Device
  • the multi -parameter chemical integrator includes a paper bridge and a chemical pellet that is included in a paper/film/foil laminated material and is sensitive to steam and temperature. When being exposed to a steam sterilization condition, the chemical pellet melts and moves along with the paper bridge to form a black stripe.
  • the black stripe may be observed in a "REJECT ACCEPT" window of the multi-parameter chemical integrator. The length of the black stripe depends on multiple parameter conditions of the sterilization device, such as steam, time, and temperature. If the black stripe can "crawl" to the side of the "ACCEPT" window, it indicates that the sterilization process is qualified.
  • the multi-parameter chemical integrator is also referred to as a "crawling card”.
  • the tested multi-parameter chemical integrator may be adhered to a test report shown in FIG. 4, and stored in a medical institution for at least more than three years. If the test card adhered to the test report can be automatically read to determine a test result, it may facilitate digital management of the test result.
  • the present disclosure is directed to provide a method and system capable of automatically reading a chemical test card, to eliminate unreliability of judgment with the naked eye, and facilitate digital management of a test result.
  • a chemical test card automatic reading method which includes: acquiring image data including a digital image of a chemical test card; extracting the digital image of the chemical test card from the acquired image data; acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data; and judging according to the feature information whether a test conducted by using the chemical test card is successful.
  • the step of extracting the digital image of the chemical test card may include conducting at least one of the following processing on the acquired image data: image enhancement processing, rotation correction, and image segmentation.
  • the chemical test card may be a BD test card, and the step of extracting the digital image of the chemical test card may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting rotation correction on the enhanced image to obtain a rotated enhanced image; and conducting image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card.
  • the step of conducting image enhancement processing may include: converting the image data from a RGB color space to a YUV color space; adjusting a pixel value of a Y ch nnel according to the following equation (1)
  • Y denotes a pixel value of the Y channel before adjustment
  • minY denotes a minimum value among pixel values of the Y channel before adjustment
  • maxY denotes a maximum value among pixel values of the Y channel before adjustment
  • Ymax denotes a maximum upper limit of the pixel value of the Y channel according to a digital level
  • Y' denotes a pixel value of the Y channel after adjustment
  • the image data is converted from the YUV color space to the RGB color space.
  • a rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to a blue-channel image.
  • the step of conducting rotation correction may include: obtaining a blue-channel image of the enhanced image; binarizing the blue-channel image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; conducting connected domain analysis to obtain a maximum connected domain; and estimating a rotation angle according to an upper boundary of the maximum connected domain, and conducting rotation correction on the enhanced image based on the rotation angle, to obtain a rotated enhanced image.
  • the step of conducting image segmentation on the rotated enhanced image may include: obtaining a blue-channel image of the rotated enhanced image; binarizing the blue-channel image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain the region of interest.
  • the step of acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data may include: obtaining a gray -channel image of an image in the region of interest including the digital image of the chemical test card; calculating a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segmenting the gray-channel image into a central region and a boundary region surrounding the central region, and calculating a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segmenting the gray-channel image into multiple blocks, and calculating a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtaining a maximum block gray average from block gray averages corresponding to blocks in the central region,
  • Adjacent blocks in the multiple blocks may overlap each other.
  • Weighted calculation may be conducted on the feature information to obtain a score for the chemical test card.
  • the score is greater than or equal to a first threshold, it may be determined that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, it may be determined that the test conducted by using the chemical test card is failed.
  • weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training.
  • the score is less than the first threshold and greater than or equal to the second threshold, it may be determined that the test conducted by using the chemical test card is in a critical state.
  • an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
  • the image data including the digital image of the chemical test card may be acquired by using a scanner.
  • step of acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data only one kind of single channel image data of the digital image of the chemical test card may be acquired, and feature information is extracted based on the kind of single channel image data.
  • the chemical test card may be a multi-parameter chemical integrator.
  • the step of extracting the digital image of the chemical test card may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conducting rotation correction on an image in the acquired region of interest.
  • the step of conducting image enhancement processing may include: conducting color space conversion on the acquired image data, wherein the color space conversion is conducted according to the following equation (2)
  • T max(0, B - R) (2) wherein, T denotes a converted color space, B denotes a blue-channel image of the image data, and R denotes a red-channel image of the image data.
  • a rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to the black-and-white binary image of the enhanced image.
  • the step of conducting image segmentation may include: binarizing the enhanced image to obtain a first black-and-white binary image; conducting connected domain analysis on the first black-and-white binary image to obtain a first maximum connected domain; estimating boundaries of the digital image including the multi -parameter chemical integrator according to the first maximum connected domain; and segmenting the enhanced image according to the boundaries, to obtain the region of interest.
  • a rotation angle may be estimated according to an upper boundary of the first maximum connected domain, and rotation correction may be conducted on the image in the region of interest based on the rotation angle.
  • the step of extracting feature information may include: binarizing the rotated image in the region of interest to obtain a second black-and-white binary image; conducting connected domain analysis on the second black-and-white binary image to obtain a second maximum connected domain; acquiring a blue-channel image of the image data, and applying estimation of boundaries and a rotation angle according to the first maximum connected domain to the blue-channel image, to obtain a corrected feature image; binarizing the corrected feature image to obtain a third black-and-white binary image; and conducting connected domain analysis on the third black-and-white binary image to obtain a third maximum connected domain.
  • the second maximum connected domain and the third maximum connected domain may be the feature information.
  • the method may further include: acquiring first identification information on an outer package of the chemical test card; acquiring second identification information on the chemical test card; and judging the first identification information and the second identification information, and conducting subsequent processing on the acquired image data when the first identification information matches the second identification information.
  • a chemical test card automatic reading system which includes: an image acquiring module, configured to acquire image data including a digital image of a chemical test card; an image processing module, configured to extract the digital image of the chemical test card from the acquired image data; a feature information extracting module, configured to acquire at least one kind of single channel image data of the digital image of the chemical test card extracted by the image processing module, and extract feature information based on the at least one kind of single channel image data; and a test result judging module, configured to judge according to the feature information extracted by the feature information extracting module whether a test conducted by using the chemical test card is successful.
  • the image processing module may be configured to conduct at least one of the following processing on the image data acquired by the image acquiring module: image enhancement processing, rotation correction, and image segmentation.
  • the chemical test card may be a BD test card
  • the image processing module may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct rotation correction on the enhanced image to obtain a rotated enhanced image; and conduct image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card.
  • the feature information extracting module may be configured to: obtain a gray-channel image of an image in the region of interest; calculate a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segment the gray-channel image into a central region and a boundary region surrounding the central region, and calculate a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segment the gray-channel image into multiple blocks, and calculate a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtain a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtain a minimum block gray average from block gray averages corresponding to blocks in the boundary region.
  • a four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average,
  • Adjacent blocks in the multiple blocks may overlap each other.
  • the test result judging module may conduct weighted calculation on the feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, the test result judging module may determine that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, the test result judging module may determine that the test conducted by using the chemical test card is failed.
  • weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training.
  • the test result judging module may determine that the test conducted by using the chemical test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
  • the image acquiring module may be a scanner.
  • the feature information extracting module may be configured to acquire only one kind of single channel image data of the digital image of the chemical test card, and extract feature information based on the kind of single channel image data.
  • the chemical test card may be a multi-parameter chemical integrator, and the image processing module may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conduct rotation correction on an image in the acquired region of interest.
  • the chemical test card automatic reading system may further include an identification information acquiring module and an identification information judging module.
  • the identification information acquiring module may be configured to acquire first identification information on an outer package of the chemical test card.
  • the image acquiring module may further acquire second identification information on the chemical test card.
  • the identification information judging module may be configured to judge the first identification information and the second identification information, and the automatic reading system conducts subsequent processing on the image data acquired by the image acquiring module when the first identification information matches the second identification information.
  • the method and system provided according to the present disclosure can automatically read a chemical test card, to eliminate unreliability of judgment with the naked eye, and facilitate digital management of a test result.
  • FIG. 1 shows a common BD test card
  • FIG. 2A to FIG. 2C show examples of a BD test card after high-pressure steam sterilization and disinfection
  • FIG. 3 shows a common multi-parameter chemical integrator
  • FIG. 4 shows a test report to which a tested multi-parameter chemical integrator is adhered
  • FIG. 5 schematically shows a flowchart of a chemical test card automatic reading method according to an exemplary embodiment
  • FIG. 6 shows an effect contrast diagram of conducting image enhancement processing on image data of a BD test card
  • FIGS. 7A-7E shows a process of conducting rotation correction processing on an enhanced image of a BD test card
  • FIG. 8A shows a process of conducting image segmentation processing on a rotated enhanced image of a BD test card
  • FIG. 8B schematically shows a region of interest of a BD test card after image segmentation processing
  • FIGS. 9A-9B shows a gray -channel image of an image in a region of interest of a BD test card and a mask image thereof;
  • FIG. 10 shows another example of image data of a BD test card
  • FIG. 1 1 shows a diagram after image enhancement processing is conducted on the image data of the multi-parameter chemical integrator shown in FIG. 4;
  • FIG. 12 shows a diagram after image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator shown in FIG. 11 ;
  • FIG. 13 shows a diagram after rotation correction is conducted on the region of interest of the multi-parameter chemical integrator shown in FIG. 12;
  • FIG. 14 shows a diagram including an "ACCEPT" window obtained based on the region of interest after rotation correction shown in FIG. 13;
  • FIG. 15 shows a diagram of a black stripe including a multi-parameter chemical integrator after image segmentation and rotation correction are conducted
  • FIG. 16 shows a diagram of a maximum connected domain including the black stripe obtained based on FIG. 15;
  • FIG. 17 schematically shows a block diagram of a chemical test card automatic reading system according to an exemplary embodiment.
  • first and second are used in this text to describe various elements; however, these elements should not be limited by these terms. These terms are merely used for distinguishing one element from another element.
  • first element may be referred to as a second element, and similarly, the second element may be referred to as a first element, without departing from the scope of the exemplary embodiments of the present disclosure.
  • the term "and/or" includes any and all combinations of one or more listed related items.
  • shown functions/actions may occur in a manner different from the order labeled in the drawing.
  • two diagrams shown successively may actually be performed basically at the same time, or sometimes may be performed in an inverse order, depending on involved
  • FIG. 5 schematically shows a flowchart of a chemical test cart automatic reading method according to an exemplary embodiment.
  • the chemical test card automatic reading method may acquire image data including a digital image of a chemical test card by using an image acquiring module (such as a scanner) (S I).
  • an image acquiring module such as a scanner
  • the present disclosure is not limited thereto, and an image including the digital image of the chemical test card may be acquired by using another imaging device according to another embodiment.
  • the digital image of the chemical test card may be extracted from the acquired image data (S2).
  • image enhancement processing, rotation correction, and image segmentation may be conducted on the acquired image data, to extract the digital image of the chemical test card.
  • the step of extracting the digital image of the chemical test card from the image data is illustrated in detail in the following by using a BD test card as an example.
  • FIG. 6 shows an effect contrast diagram of conducting image enhancement processing on image data of a BD test card.
  • the diagram at a left part is original image data of the BD test card.
  • the original image data is not suitable for being directly used for the operation of extracting feature information, due to reasons such as the batch of the test card, the type of a reactant, and parameter setting of an imaging device. Therefore, image enhancement processing needs to be conducted on the original image data. It is judged whether a BD test is qualified by considering the characteristic of a BD test, that is, according to whether the color of the reactant is changed uniformly. Therefore, the image enhancement processing conducted on the image data of the BD test card should not affect the color uniformity of a stripe image of the reactant.
  • a manner of conducting image enhancement processing on the image data of the BD test card may be provided as: converting the image data from a RGB color space to a YUV color space; adjusting a pixel value of a Y channel accordin to the following equation (1) (1) wherein, Y denotes a pixel value of the Y channel before adjustment, minY denotes a minimum value among pixel values of the Y channel before adjustment, maxY denotes a maximum value among pixel values of the Y channel before adjustment, Ymax denotes a maximum upper limit of the pixel value of the Y channel according to a digital level, and Y' denotes a pixel value of the Y channel after adjustment; and converting the image data from the YUV color space to the RGB color space based on the pixel value of the Y channel after adjustment.
  • the brightness of the image data may be extended to the whole range thereof in accordance with a digital level without affect the hue of the image data.
  • the diagram at the right part of FIG. 6 is an enhanced image of the BD test card after the image enhancement processing. It can be seen that, the contrast of the enhanced image is increased as compared with the contrast of the original image data, and this facilitates various types of subsequent image processing.
  • FIGS. 7A-7E shows a process of conducting rotation correction processing on an enhanced image of a BD test card.
  • a blue channel is more advantageous to distinguish a stripe region from background information of the BD test card, and therefore, a blue-channel image is first extracted for the enhanced image of the BD test card ( FIG. 7A).
  • the blue-channel image thereof may be extracted after the size of the image may be reduced (e.g., reduced to 1/64 of the original image size), to reduce the overhead of calculation.
  • a fixed threshold 128 is used to conduct a binarization processing, to obtain a black-and-white binary image ( FIG. 7B). For example, in a case of 8-bit data, pixels of which blue component values are less than 128 are defined as white pixels; otherwise, the pixels are defined as black pixels.
  • FIG. 7C A morphological operation is used for the black-and-white binary image of the BD test card to remove noises, and stripe regions are joined into a complete connected domain ( FIG. 7C).
  • a connected domain analysis is conducted to find a maximum connected domain ( FIG. 7D), and at the same time, a rotation angle ⁇ is estimated according to an upper boundary of the connected domain.
  • Selection correction is conducted based on the rotation angle ⁇ , to obtain a corrected result ( FIG. 7E).
  • rotation correction is conducted on the enhanced image of the BD test card based on the rotation angle ⁇ , to obtain a rotated enhanced image of the BD test card.
  • the above process may be repeated multiple times, to obtain a precise positioning result of the BD test card.
  • FIG. 8A shows a process of conducting image segmentation processing on the rotated enhanced image of a BD test card
  • FIG. 8B schematically shows a region of interest (ROI) of the BD test card after image segmentation processing.
  • ROI region of interest
  • the step of conducting image segmentation on the rotated enhanced image of the BD test card includes:
  • the "foreground pixel" described in this text is such a pixel that in a certain region, the brightness of the foreground pixel is less than the average brightness of an image in this region.
  • a reactant after disinfection processing assume apparent dark brown to black, and the brightness thereof is less than the brightness of other parts of the image data.
  • image enhancement processing is conducted on the image data such that the brightness distribution thereof is extended to the whole threshold range.
  • the blue-channel image of the enhanced image is extracted, and therefore, dark stripe parts are obviously distinguished from other parts of the image when counting statistics is conducted for the foreground pixels.
  • boundaries of the ROI may be determined by simply applying a threshold method. For example, when a difference between two adjacent count values of the foreground pixels is greater than a predetermined threshold, boundaries of the ROI may be determined.
  • the step of conducting image segmentation on the rotated enhanced image may include: obtaining a blue-channel image of the rotated enhanced image; binarizing the blue-channel image of the rotated enhanced image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain the ROI.
  • the image processing process is similar to the image processing process described with reference to FIGS. 7A-7E, and therefore, repeated descriptions thereof are omitted herein.
  • the above image processing is conducted based on the rotated image, and therefore, it is unnecessary to determine a rotation angle by using the obtained connected domain. Instead, boundaries of the ROI may be determined by using the obtained connected domain, and image segmentation is conducted on the rotated enhanced image by using the boundaries of the ROI, to obtain the ROI.
  • FIGS. 9A-9B shows a gray-channel image of an image in a ROI of a BD test card and a mask image thereof.
  • the step of extracting feature information may include: obtaining a gray-channel image of an image in the ROI (as shown in FIG. 9A); calculating a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the ROI; segmenting the gray-channel image into a central region and a boundary region surrounding the central region, and calculating a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segmenting the gray-channel image into multiple blocks, and calculating a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtaining a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtaining a minimum block gray average from block gray averages corresponding to blocks in the boundary region.
  • a four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average is the feature information.
  • a mask image may be constructed based on the ROI of the BD test card (as shown in FIG. 9B).
  • the foreground pixels are displayed as white pixels (or marked as "1 "), and other parts are displayed as black pixels (or marked as "0").
  • a block segmentation manner of a gray-channel image and a segmentation manner of the central region are applied to the mask image, and the quantities (or count sums) of white pixels in each block, in the central region, and in the whole image are counted statistically respectively, so as to obtain the quantities of foreground pixels in each block, in the central region, and in the whole image.
  • the size (a dashed rectangular part) of a block for segmenting the gray-channel image is exemplified in FIG. 9A, but the present disclosure is not limited thereto, and the size of the block may be adjusted according to an actual application situation.
  • an overlapping portion may exist between adjacent blocks among the multiple blocks.
  • Such a block segmentation manner can ensure that the calculated gray average has a desirable partial characteristic, and at the same time is more stable than the gray value of a single pixel.
  • the segmented central region is exemplified in FIG. 9B, but the present disclosure is not limited thereto, and the size of the central region may be adjusted according to an actual application situation. It should be realized that, the same block segmentation manner and central region segmentation manner are applied to the gray -channel image and the mask image, and therefore, the block segmentation manner shown in the gray -channel image will be applied to the mask image, and the central region segmentation manner shown in the mask image will be applied to the gray-channel image.
  • the maximum block gray average is generally extracted from a block located at the central region
  • the minimum gray average is generally extracted from a block located at the boundary region. This is because that in the process of disinfection, high-temperature steam permeates from edges of the BD test card to the center, and a reactant located at the edge portion will change its color in the first place, and the color thereof is the most fully changed.
  • the reactant may change its color from dark brown to black, and therefore, the reactant having the color fully changed may have a relatively low gray value.
  • the minimum gray average acquired from the boundary region may be considered as the part of the BD test card where the reaction is the most fully conducted, and the maximum gray average acquired from the central region may be considered as the part of the BD test card where the reaction is the least fully conducted.
  • the maximum gray average and the minimum gray average introduced into the feature information may reflect the uniformity of color change of the BD test card.
  • the calculation of the gray average of each part depends on the quantity of foreground pixels of the corresponding part.
  • the foreground pixels correspond to the dark stripe part of the BD test card (that is, a reactant region on the BD test card).
  • the gray average of each part is calculated, merely gray values of pixels corresponding to the reactant rather than pixels corresponding to the blank region (that is, a white stripe part) are considered.
  • the feature information may be extracted based on gray-channel image data and yellow-channel image data of the digital image of the chemical test card. For example, a full-image gray average and a central gray average may be calculated based on the yellow-channel image data, and the maximum block gray average and the minimum block gray average are calculated based on the gray-channel image.
  • weighted calculation may be conducted on the extracted feature information to obtain a score for the chemical test card.
  • a score is greater than or equal to a first threshold, it may be determined that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, it may be determined that the test conducted by using the chemical test card is failed.
  • the score is less than the first threshold and greater than or equal to the second threshold, it may be determined that the test conducted by using the chemical test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
  • the expert includes (but is not limited to) a person who is very experienced in judging the BD test card, for example, a head nurse in a medical institution.
  • various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training. The acquisition of samples also depends on introduction of expert opinions.
  • test samples in different cases may be obtained in advance according to judgment results of experienced experts, so as to conduct training to obtain various weights in the weighted calculation as well as the first threshold and the second threshold that are used for determining the extracted feature information.
  • the expert opinion may be further introduced to a test result in a critical state (that is, between the first threshold and the second threshold), and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion, to constantly perfect the whole automatic reading method, such that the result of automatic reading may be better consistent with the expert opinion.
  • FIG. 10 shows another example of image data of a BD test card.
  • the image data of the BD test card includes a stripe region of the BD test card and an identification information region near the stripe region.
  • the identification information region is shown as a QR code.
  • only the upper half part of the image data of the BD test card is shown in the example shown in FIG. 10, and the whole image data further includes a lower part having an area almost the same as that of the shown part.
  • the present disclosure provides a process of modified image processing.
  • the process of modified image processing may include : obtaining a blue-channel image of the image data of the BD test card and reducing the image appropriately; binarizing the obtained image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain a maximum connected domain.
  • the maximum connected domain may cover the stripe region of the BD test card.
  • the process of modified image processing may further include: amplifying boundaries of the maximum connected domain proportionally, to cover an identification information region; and conducting image segmentation on the image data of the BD test card according to the amplified boundaries, to obtain a region of interest.
  • the region of interest not only includes the stripe region of the BD test card, but also includes the identification information region.
  • the process of modified image processing may further include: estimating a rotation angle of the image according to the boundaries of the maximum connection domain obtained previously (before amplification); and conducting rotation correction on the region of interest including the stripe region of the BD test card and the identification information region by using the obtained rotation angle, to obtain a rotated region of interest.
  • the rotation angle of the image is estimated according to the boundaries of the maximum connected domain, and boundaries of the region of interest are obtained according to the boundaries of the amplified maximum connected domain.
  • Image segmentation is conducted on the image data of the BD test card by using the boundaries of the region of interest, and then rotation correction is conducted by using the rotation angle on the region of interest obtained after segmentation.
  • conducting an operation of image segmentation before rotation correction may effectively reduce the overhead of calculation amount.
  • the rotated region of interest includes the stripe region of the BD test card and the identification information region.
  • a digital image of the stripe region of the BD test card may be acquired by using the manner described in the above embodiment, and feature information is extracted.
  • identification information of the BD test card may further be obtained by using a digital image of the identification information region.
  • the BD test card may be authenticated by using the identification information of the BD test card.
  • acquiring identification information on an outer package of the chemical test card that is, first identification information
  • the BD test card may be authenticated by using the first identification information and identification information of the BD test card (that is, second identification information).
  • Subsequent processing e.g., extracting feature information
  • Subsequent processing is conducted on the acquired image data when the first identification information matches the second identification information.
  • no subsequent processing is conducted when the first identification information does not match the second identification information, and it is prompted to scan the chemical test card again.
  • the first identification information and the second identification information may be different from each other but have a unique corresponding relationship. In this way, test cards are prevented from being mixed up.
  • the step of extracting the digital image of the chemical test card from the image data is illustrated in detail in the following by using a multi-parameter chemical integrator as an example.
  • multi-parameter chemical integrator adhered to a test report and determining a test result may facilitate digital management of the test result.
  • an area of image data occupied by the multi-parameter chemical integrator is relatively small, and therefore, an operation of image segmentation may be conducted first before rotation correction in the process of extracting the digital image of the chemical test card, to reduce the overhead of the calculation amount.
  • the step of extracting the digital image of the chemical test card (S2) shown in FIG. 5 may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting image segmentation on the enhanced image to obtain a ROI including the digital image of the chemical test card; and conducting rotation correction on an image in the acquired ROI.
  • FIG. 11 shows a diagram after image enhancement processing is conducted on the image data of the multi-parameter chemical integrator shown in FIG. 4;
  • FIG. 12 shows a diagram after image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator shown in FIG. 11 ;
  • FIG. 13 shows a diagram after rotation correction is conducted on a ROI of the multi-parameter chemical integrator shown in FIG. 12;
  • FIG. 14 shows a diagram including an "ACCEPT" window obtained based on the ROI after rotation correction shown in FIG. 13;
  • FIG. 15 shows a diagram of a black stripe including a multi-parameter chemical integrator after image segmentation and rotation correction are conducted; and
  • FIG. 16 shows a diagram of a maximum connected domain including the black stripe obtained based on FIG. 15.
  • the step of conducting image enhancement processing on the image data of the multi-parameter chemical integrator may include: conducting color space conversion according to the following equation (2)
  • T max(0, B-R) (2) wherein, T denotes a converted color space, and B denotes a blue-channel image of image data of the multi -parameter chemical integrator, and R denotes a red-channel image of image data of the multi -parameter chemical integrator.
  • the multi-parameter chemical integrator has a white bottom printed with a turquoise pattern.
  • an "ACCEPT" window of a PCD card is printed as a turquoise pattern having a large area. Therefore, the enhanced image after the image enhancement processing is conducive to subsequent determination of the position of the "ACCEPT" window.
  • a rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to the black-and-white binary image of the enhanced image (that is, the image shown in FIG. 11) of the multi -parameter chemical integrator.
  • the step of conducting image segmentation on the enhanced image of the multi-parameter chemical integrator may include: binarizing the enhanced image of the multi-parameter chemical integrator to obtain a first black-and-white binary image.
  • the "ACCEPT" window that originally assumes turquoise may have a large contrast compared with display content of other parts. Therefore, after the binarization processing (that is, pixels of which brightness is greater than a threshold are defined as white pixels, and otherwise, pixels are defined as black pixels), the "ACCEPT" window displayed as white pixels may be clearly embodied in the obtained first black-and-white binary image.
  • a first maximum connected domain including the "ACCEPT" window may be obtained through connected domain analysis.
  • the position of the "ACCEPT" window on the multi-parameter chemical integrator is relatively fixed, and therefore, boundaries of the digital image including the multi-parameter chemical integrator may be estimated by using the obtained first maximum connected domain, and image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator according to the estimated boundaries to obtain a ROI.
  • a rotation angle may be estimated according to an upper boundary of the first maximum connected domain, and rotation correction may be conducted on the image (as shown in FIG. 12) in the ROI of the multi-parameter chemical integrator based on the rotation angle.
  • the step of extracting feature information (S3) shown in FIG. 5 may include: binarizing the rotated image (as shown in FIG. 13) in the ROI to obtain a second black-and-white binary image; and conducting connected domain analysis on the second black-and-white binary image, to obtain a second maximum connected domain (as shown by the dashed block in FIG. 14).
  • the "ACCEPT" window included in the second maximum connected domain may have a basically vertical left boundary, thus facilitating subsequent judgment on whether a black stripe on the
  • the step of extracting feature information (S3) shown in FIG. 5 may further include a step of extracting a third maximum connected domain including the black stripe on the multi-parameter chemical integrator.
  • the step may include: acquiring a blue-channel image of the image data of the multi-parameter chemical integrator, and applying estimation of boundaries and a rotation angle according to the first maximum connected domain to the blue-channel image, to obtain a corrected feature image (as shown in FIG. 15); binarizing the corrected feature image to obtain a third black-and-white binary image; and conducting connected domain analysis on the third black-and-white binary image to obtain a third maximum connected domain (as shown by the dashed block in FIG. 16).
  • the third black-and-white binary image and the third maximum connected domain based on the third black-and-white binary image are acquired from the image after segmentation and rotation. Therefore, the black stripe included in the third maximum connected domain may have a basically horizontal shape.
  • the obtained second maximum connected domain and the third maximum connected domain are feature information extracted in the step of extracting feature information (S3) shown in the drawing. Subsequently, it may be judged according to the extracted feature information whether a test conducted by using the multi-parameter chemical integrator is successful (S4). Specifically, according to the embodiment of the present disclosure, when it is determined according to the extracted feature that the second maximum connected domain and the third maximum connected domain have an overlapping portion, it may be determined that the test conducted by using the multi-parameter chemical integrator is successful (that is, whether the black stripe "crawls" to the side of the "ACCEPT" window); otherwise, it may be determined that the test conducted by using the multi-parameter chemical integrator is failed.
  • the chemical test card is the multi-parameter chemical integrator by using a "crawling card" as an example, but the present disclosure is not limited thereto. It should be realized that the chemical test card automatic reading method according to the present disclosure may be applied to various multi-parameter chemical integrators used in various cases such as steam sterilization, ethylene oxide sterilization, and hydrogen peroxide sterilization.
  • FIG. 17 schematically shows a block diagram of a chemical test card automatic reading system according to an exemplary embodiment.
  • an automatic reading system 1000 may include an image acquiring module 1001, an image processing module 1002, a feature information extracting module 1003, and a test result judging module 1004.
  • the image acquiring module 1001 is configured to acquire image data including a digital image of a chemical test card.
  • the image processing module 1002 is configured to extract the digital image of the chemical test card from the acquired image data.
  • the feature information extracting module 1003 is configured to acquire at least one kind of single channel image data of the digital image of the chemical test card extracted by the image processing module 1002, and extract feature information based on the at least one kind of single channel image data.
  • the test result judging module 1004 is configured to judge according to the feature information extracted by the feature information extracting module 1003 whether a test conducted by using the chemical test card is successful.
  • the image processing module 1002 may be configured to conduct image enhancement processing, rotation correction, and image segmentation on the image data acquired by the image acquiring module 1001, to extract the digital image of the chemical test card.
  • the image processing module 1002 may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct rotation correction on the enhanced image to obtain a rotated enhanced image; and conduct image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card.
  • the feature information extracting module 1003 may be configured to: obtain a gray-channel image of an image in the region of interest; calculate a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segment the gray-channel image into a central region and a boundary region surrounding the central region, and calculate a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segment the gray-channel image into multiple blocks, and calculate a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtain a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtain a minimum block gray average from block gray averages corresponding to blocks in the boundary region.
  • a four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray
  • the test result judging module 1004 may conduct weighted calculation on the feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is successful, and when the score is less than a second threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is failed. Various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training. When the score is less than the first threshold and greater than or equal to the second threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
  • the image processing module 1002 may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conduct rotation correction on an image in the acquired region of interest.
  • the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 may be implemented as program modules running on a computer, or may be implemented as dedicated (or universal) hardware modules that can execute corresponding functions.
  • the image acquiring module 1001, the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 are shown as modules independent of each other; however, according to the embodiment of the present disclosure, these modules 1001 to 1004 may be all or partially implemented by using a single module. For example, a dedicated device having a scanning function may be implemented to implement all or partial functions of the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004.
  • the function of the image processing module 1002 may be specifically implemented by multiple sub-modules.
  • the automatic reading system 1000 may further include a storing module configured to store, for example, software program codes for implementing the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 and/or the first threshold and the second threshold for judging the score for the chemical test card.
  • a storing module configured to store, for example, software program codes for implementing the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 and/or the first threshold and the second threshold for judging the score for the chemical test card.
  • the automatic reading system 1000 may further include a displaying module configured to display, for example, a judgment result output by the test result judging module 1004 and/or display, for example, prompt information that an expert opinion needs to be provided when the test is in a critical state.
  • a displaying module configured to display, for example, a judgment result output by the test result judging module 1004 and/or display, for example, prompt information that an expert opinion needs to be provided when the test is in a critical state.
  • the automatic reading system 1000 may be implemented as a system including (but not limited to) a scanner and a desktop computer. However, the present disclosure is not limited thereto, and according to another embodiment of the present disclosure, the automatic reading system 1000 may be implemented as a system including (but not limited to) a scanner and a portable mobile device, or the automatic reading system 1000 may be implemented as a dedicated device having a scanning function. Moreover, the automatic reading system 1000 may further include one or more modules and/or devices configured to implement various other additional functions.
  • the automatic reading system 1000 may further include an identification information acquiring module 1005 and an identification information judging module 1006.
  • the identification information acquiring module 1005 is configured to acquire first identification information on an outer package of the chemical test card.
  • the identification information acquiring module 1005 may be implemented as a bar code reading device, configured to read bar code or QR code information printed on the outer package of the chemical test card, to acquire the first identification information.
  • the image processing module 1002 may first extract second identification information from the digital data of the chemical test card.
  • the second identification information may be implemented as a bar code or a QR code (referring to an example shown in FIG. 10) printed on the chemical test card (or printed on a test report including the chemical test card).
  • the identification information judging module 1006 judges the first identification information and the second identification information.
  • a subsequent processing (for example, extracting feature information by using the feature information extracting module 1003) may be conducted when the first identification information matches the second identification information.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Animal Behavior & Ethology (AREA)
  • Veterinary Medicine (AREA)
  • Public Health (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Epidemiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

Embodiments of the present disclosure provide a chemical test card automatic reading method and system. The automatic reading method includes: acquiring image data including a digital image of a chemical test card; extracting the digital image of the chemical test card from the acquired image data; acquiring at least one kind of single channel image data of the digital image of the chemical test card; extracting feature information based on the at least one kind of single channel image data; and judging according to the feature information whether a test conducted by using the chemical test card is successful.

Description

CHEMICAL TEST CARD AUTOMATIC READING METHOD AND SYSTEM
TECHNICAL FIELD
The present disclosure relates to the technical field of image recognition, and in particular, to a chemical test card automatic reading method and system.
BACKGROUND
In the medical and health care industry, a high-pressure steam sterilization device is generally used for disinfecting and sterilizing medical instruments. In the process of disinfection, if air in the sterilization device is not completely discharged or air leakage occurs, some air may reside in the sterilization device. The residual air may affect permeation of high-pressure steam to a to-be-disinfected article in the sterilization device in a steam exposure stage of the process of sterilization, thus reducing the sterilization effect.
A BD (Bowie-Dick) test is named after last names of inventors thereof, i.e., J. H. Bowie and J. Dick. The BD test is used to evaluate the vacuum state of a sterilization device, thereby judging whether the sterilization device works normally. Before the high-pressure steam sterilization and disinfection is performed, a BD test card in a BD test pack is placed into a sterilization device together with a to-be-disinfected article, to test the sterilization device.
FIG. 1 shows a common BD test card, and FIG. 2A to FIG. 2C show examples of a BD test card after high-pressure steam sterilization and disinfection.
As shown in FIG. 1, a stripe region of the BD test card mainly includes two parts, namely, a reactant region and a blank region. A change generated on the BD test card after being heated by a high temperature is generally the color change of the reactant region. Specifically, the reactant on the BD test card is originally in a light yellow color, and the reactant may change from light yellow to uniform dark brown and to black after being heated by a high temperature, while the color of the blank region is always white. If the color of the reactant is changed uniformly, as shown in FIG. 2A, it indicates that the BD test is qualified, and the sterilization device can work normally. If the color of the reactant of the BD test card is changed non-uniformly, as shown in FIG. 2B, it indicates that the BD test is unqualified, and the sterilization device has an air leakage phenomenon and cannot work normally.
In an existing sterilization quality control system, generally an operator judges whether the color change effect of a BD test card reaches a designated standard with the naked eye, and such a judging method is very subjective. Different operators have different sensitivities on changes in color and brightness, and this may result in different judgment results of different operators for the same BD test card. Particularly, for an example in a critical state as shown in FIG. 2C, different operators may obtain different judgment results. Moreover, the same operator may also obtain different judgment results for the same BD test card in different time. This causes unreliability of judging the color change effect of the BD test card by naked eyes, thus bringing about potential risks for medical and health care quality management.
A Process Challenge Device (PCD) test is another common test method for testing a sterilization device in the medical and health care industry. FIG. 3 shows a common
multi -parameter chemical integrator. As shown in FIG. 3, the multi -parameter chemical integrator includes a paper bridge and a chemical pellet that is included in a paper/film/foil laminated material and is sensitive to steam and temperature. When being exposed to a steam sterilization condition, the chemical pellet melts and moves along with the paper bridge to form a black stripe. The black stripe may be observed in a "REJECT ACCEPT" window of the multi-parameter chemical integrator. The length of the black stripe depends on multiple parameter conditions of the sterilization device, such as steam, time, and temperature. If the black stripe can "crawl" to the side of the "ACCEPT" window, it indicates that the sterilization process is qualified. If the black stripe does not enter the "ACCEPT" window, but stays at the side of the "REJECT" window, it indicates that a to-be-disinfected article cannot be fully exposed to the steam sterilization environment, and the sterilization process is unqualified. Therefore, the multi-parameter chemical integrator is also referred to as a "crawling card".
Generally, the tested multi-parameter chemical integrator may be adhered to a test report shown in FIG. 4, and stored in a medical institution for at least more than three years. If the test card adhered to the test report can be automatically read to determine a test result, it may facilitate digital management of the test result.
SUMMARY
The present disclosure is directed to provide a method and system capable of automatically reading a chemical test card, to eliminate unreliability of judgment with the naked eye, and facilitate digital management of a test result.
According to one aspect of the present disclosure, a chemical test card automatic reading method is provided, which includes: acquiring image data including a digital image of a chemical test card; extracting the digital image of the chemical test card from the acquired image data; acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data; and judging according to the feature information whether a test conducted by using the chemical test card is successful.
The step of extracting the digital image of the chemical test card may include conducting at least one of the following processing on the acquired image data: image enhancement processing, rotation correction, and image segmentation. The chemical test card may be a BD test card, and the step of extracting the digital image of the chemical test card may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting rotation correction on the enhanced image to obtain a rotated enhanced image; and conducting image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card.
The step of conducting image enhancement processing may include: converting the image data from a RGB color space to a YUV color space; adjusting a pixel value of a Y ch nnel according to the following equation (1)
Figure imgf000005_0001
wherein, Y denotes a pixel value of the Y channel before adjustment, minY denotes a minimum value among pixel values of the Y channel before adjustment, maxY denotes a maximum value among pixel values of the Y channel before adjustment, Ymax denotes a maximum upper limit of the pixel value of the Y channel according to a digital level, and Y' denotes a pixel value of the Y channel after adjustment; and converting the image data from the YUV color space to the RGB color space based on the pixel value of the Y channel after adjustment.
The image data is converted from the YUV color space to the RGB color space. A rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to a blue-channel image.
The step of conducting rotation correction may include: obtaining a blue-channel image of the enhanced image; binarizing the blue-channel image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; conducting connected domain analysis to obtain a maximum connected domain; and estimating a rotation angle according to an upper boundary of the maximum connected domain, and conducting rotation correction on the enhanced image based on the rotation angle, to obtain a rotated enhanced image.
The step of conducting image segmentation on the rotated enhanced image may include: obtaining a blue-channel image of the rotated enhanced image; binarizing the blue-channel image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain the region of interest.
The step of acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data may include: obtaining a gray -channel image of an image in the region of interest including the digital image of the chemical test card; calculating a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segmenting the gray-channel image into a central region and a boundary region surrounding the central region, and calculating a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segmenting the gray-channel image into multiple blocks, and calculating a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtaining a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtaining a minimum block gray average from block gray averages corresponding to blocks in the boundary region. A four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average may be the feature information.
Adjacent blocks in the multiple blocks may overlap each other.
Weighted calculation may be conducted on the feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, it may be determined that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, it may be determined that the test conducted by using the chemical test card is failed.
Various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training.
When the score is less than the first threshold and greater than or equal to the second threshold, it may be determined that the test conducted by using the chemical test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
The image data including the digital image of the chemical test card may be acquired by using a scanner.
In the step of acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data, only one kind of single channel image data of the digital image of the chemical test card may be acquired, and feature information is extracted based on the kind of single channel image data.
The chemical test card may be a multi-parameter chemical integrator. The step of extracting the digital image of the chemical test card may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conducting rotation correction on an image in the acquired region of interest.
The step of conducting image enhancement processing may include: conducting color space conversion on the acquired image data, wherein the color space conversion is conducted according to the following equation (2)
T = max(0, B - R) (2) wherein, T denotes a converted color space, B denotes a blue-channel image of the image data, and R denotes a red-channel image of the image data.
A rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to the black-and-white binary image of the enhanced image.
The step of conducting image segmentation may include: binarizing the enhanced image to obtain a first black-and-white binary image; conducting connected domain analysis on the first black-and-white binary image to obtain a first maximum connected domain; estimating boundaries of the digital image including the multi -parameter chemical integrator according to the first maximum connected domain; and segmenting the enhanced image according to the boundaries, to obtain the region of interest.
A rotation angle may be estimated according to an upper boundary of the first maximum connected domain, and rotation correction may be conducted on the image in the region of interest based on the rotation angle.
The step of extracting feature information may include: binarizing the rotated image in the region of interest to obtain a second black-and-white binary image; conducting connected domain analysis on the second black-and-white binary image to obtain a second maximum connected domain; acquiring a blue-channel image of the image data, and applying estimation of boundaries and a rotation angle according to the first maximum connected domain to the blue-channel image, to obtain a corrected feature image; binarizing the corrected feature image to obtain a third black-and-white binary image; and conducting connected domain analysis on the third black-and-white binary image to obtain a third maximum connected domain. The second maximum connected domain and the third maximum connected domain may be the feature information.
When it is determined according to the extracted feature information that the second maximum connected domain and the third maximum connected domain have an overlapping portion, it may be determined that the test conducted by using the chemical test card is successful; otherwise, it may be determined that the test conducted by using the chemical test card is failed. Before the step of extracting the digital image of the chemical test card, the method may further include: acquiring first identification information on an outer package of the chemical test card; acquiring second identification information on the chemical test card; and judging the first identification information and the second identification information, and conducting subsequent processing on the acquired image data when the first identification information matches the second identification information.
According to another aspect of the present disclosure, a chemical test card automatic reading system is provided, which includes: an image acquiring module, configured to acquire image data including a digital image of a chemical test card; an image processing module, configured to extract the digital image of the chemical test card from the acquired image data; a feature information extracting module, configured to acquire at least one kind of single channel image data of the digital image of the chemical test card extracted by the image processing module, and extract feature information based on the at least one kind of single channel image data; and a test result judging module, configured to judge according to the feature information extracted by the feature information extracting module whether a test conducted by using the chemical test card is successful.
The image processing module may be configured to conduct at least one of the following processing on the image data acquired by the image acquiring module: image enhancement processing, rotation correction, and image segmentation.
The chemical test card may be a BD test card, and the image processing module may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct rotation correction on the enhanced image to obtain a rotated enhanced image; and conduct image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card.
The feature information extracting module may be configured to: obtain a gray-channel image of an image in the region of interest; calculate a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segment the gray-channel image into a central region and a boundary region surrounding the central region, and calculate a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segment the gray-channel image into multiple blocks, and calculate a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtain a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtain a minimum block gray average from block gray averages corresponding to blocks in the boundary region. A four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average may be the feature information.
Adjacent blocks in the multiple blocks may overlap each other.
The test result judging module may conduct weighted calculation on the feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, the test result judging module may determine that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, the test result judging module may determine that the test conducted by using the chemical test card is failed.
Various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training.
When the score is less than the first threshold and greater than or equal to the second threshold, the test result judging module may determine that the test conducted by using the chemical test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
The image acquiring module may be a scanner.
The feature information extracting module may be configured to acquire only one kind of single channel image data of the digital image of the chemical test card, and extract feature information based on the kind of single channel image data.
The chemical test card may be a multi-parameter chemical integrator, and the image processing module may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conduct rotation correction on an image in the acquired region of interest.
The chemical test card automatic reading system may further include an identification information acquiring module and an identification information judging module. The identification information acquiring module may be configured to acquire first identification information on an outer package of the chemical test card. The image acquiring module may further acquire second identification information on the chemical test card. The identification information judging module may be configured to judge the first identification information and the second identification information, and the automatic reading system conducts subsequent processing on the image data acquired by the image acquiring module when the first identification information matches the second identification information.
The method and system provided according to the present disclosure can automatically read a chemical test card, to eliminate unreliability of judgment with the naked eye, and facilitate digital management of a test result. BRIEF DESCRIPTION OF DRAWINGS
Exemplary embodiments are described in detail with reference to accompanying drawings, to further clarify the above and other features and advantages of the exemplary embodiments of the present disclosure. The accompanying drawings are aimed at describing the exemplary embodiments of the present disclosure, and should not be construed as limiting the expected range of claims. In the accompanying drawings:
FIG. 1 shows a common BD test card;
FIG. 2A to FIG. 2C show examples of a BD test card after high-pressure steam sterilization and disinfection;
FIG. 3 shows a common multi-parameter chemical integrator;
FIG. 4 shows a test report to which a tested multi-parameter chemical integrator is adhered;
FIG. 5 schematically shows a flowchart of a chemical test card automatic reading method according to an exemplary embodiment;
FIG. 6 shows an effect contrast diagram of conducting image enhancement processing on image data of a BD test card;
FIGS. 7A-7E shows a process of conducting rotation correction processing on an enhanced image of a BD test card;
FIG. 8A shows a process of conducting image segmentation processing on a rotated enhanced image of a BD test card;
FIG. 8B schematically shows a region of interest of a BD test card after image segmentation processing;
FIGS. 9A-9B shows a gray -channel image of an image in a region of interest of a BD test card and a mask image thereof;
FIG. 10 shows another example of image data of a BD test card;
FIG. 1 1 shows a diagram after image enhancement processing is conducted on the image data of the multi-parameter chemical integrator shown in FIG. 4;
FIG. 12 shows a diagram after image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator shown in FIG. 11 ;
FIG. 13 shows a diagram after rotation correction is conducted on the region of interest of the multi-parameter chemical integrator shown in FIG. 12;
FIG. 14 shows a diagram including an "ACCEPT" window obtained based on the region of interest after rotation correction shown in FIG. 13;
FIG. 15 shows a diagram of a black stripe including a multi-parameter chemical integrator after image segmentation and rotation correction are conducted; FIG. 16 shows a diagram of a maximum connected domain including the black stripe obtained based on FIG. 15; and
FIG. 17 schematically shows a block diagram of a chemical test card automatic reading system according to an exemplary embodiment.
DESCRIPTION OF EMBODIMENTS
The exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. However, the present disclosure may be exemplified according to many different forms, and should not be construed as being limited to the exemplary embodiments described in this text. Moreover, these embodiments are provided so that the present disclosure is thorough and complete, and fully conveys the scope of the present disclosure to those skilled in the art.
In the accompanying drawings, shapes and sizes of elements may be amplified for clarity, and identical reference numerals will always be used to indicate identical or similar elements.
It should be understood that, terms such as first and second are used in this text to describe various elements; however, these elements should not be limited by these terms. These terms are merely used for distinguishing one element from another element. For example, the first element may be referred to as a second element, and similarly, the second element may be referred to as a first element, without departing from the scope of the exemplary embodiments of the present disclosure. As used in this text, the term "and/or" includes any and all combinations of one or more listed related items.
Terms used in this text are merely used for describing the exemplary embodiments illustrated in this text, and are not intended to limit the present disclosure. As used in this text, singular forms "a", "an" and "the" are also intended to include plural forms, unless otherwise specified in the context. It should be further understood that, when terms "include", "provided with" and/or "have" are used in the exemplary embodiments, they indicate existence of the feature, entirety, step, operation, element and/or member, but do not exclude existence or addition of other one or more features, entireties, steps, operations, elements, members and/or combinations thereof.
It should be further noted that, in some alternative implementations, shown functions/actions may occur in a manner different from the order labeled in the drawing. For example, two diagrams shown successively may actually be performed basically at the same time, or sometimes may be performed in an inverse order, depending on involved
functions/actions .
Unless otherwise specified, all terms including illustrative terms or technical terms used in this text should be understood as having meanings apparent to those of ordinary skill in the art. In addition, a term defined in an ordinary dictionary and used in the following description should be understood as having a meaning equivalent to the meaning used in the related description. Moreover, the term should not be understood as being ideal or excessively official, unless otherwise specified in this text.
FIG. 5 schematically shows a flowchart of a chemical test cart automatic reading method according to an exemplary embodiment.
As shown in FIG. 5, the chemical test card automatic reading method according to an embodiment of the present disclosure may acquire image data including a digital image of a chemical test card by using an image acquiring module (such as a scanner) (S I). However, the present disclosure is not limited thereto, and an image including the digital image of the chemical test card may be acquired by using another imaging device according to another embodiment.
After the image data is acquired, the digital image of the chemical test card may be extracted from the acquired image data (S2). According to the embodiment of the present disclosure, in the process of extracting the digital image of the chemical test card, image enhancement processing, rotation correction, and image segmentation may be conducted on the acquired image data, to extract the digital image of the chemical test card.
Exemplary embodiment: BD test card
The step of extracting the digital image of the chemical test card from the image data is illustrated in detail in the following by using a BD test card as an example.
FIG. 6 shows an effect contrast diagram of conducting image enhancement processing on image data of a BD test card.
Referring to FIG. 6, the diagram at a left part is original image data of the BD test card. The original image data is not suitable for being directly used for the operation of extracting feature information, due to reasons such as the batch of the test card, the type of a reactant, and parameter setting of an imaging device. Therefore, image enhancement processing needs to be conducted on the original image data. It is judged whether a BD test is qualified by considering the characteristic of a BD test, that is, according to whether the color of the reactant is changed uniformly. Therefore, the image enhancement processing conducted on the image data of the BD test card should not affect the color uniformity of a stripe image of the reactant.
According to an embodiment of the present disclosure, a manner of conducting image enhancement processing on the image data of the BD test card may be provided as: converting the image data from a RGB color space to a YUV color space; adjusting a pixel value of a Y channel accordin to the following equation (1)
Figure imgf000012_0001
(1) wherein, Y denotes a pixel value of the Y channel before adjustment, minY denotes a minimum value among pixel values of the Y channel before adjustment, maxY denotes a maximum value among pixel values of the Y channel before adjustment, Ymax denotes a maximum upper limit of the pixel value of the Y channel according to a digital level, and Y' denotes a pixel value of the Y channel after adjustment; and converting the image data from the YUV color space to the RGB color space based on the pixel value of the Y channel after adjustment.
By means of the above image enhancement processing, the brightness of the image data may be extended to the whole range thereof in accordance with a digital level without affect the hue of the image data. For example, in a case of 8-bit data, a digital level thereof is 28=256, and the brightness of the image may be extended to the whole range of [0, 255] by using the above image enhancement processing.
The diagram at the right part of FIG. 6 is an enhanced image of the BD test card after the image enhancement processing. It can be seen that, the contrast of the enhanced image is increased as compared with the contrast of the original image data, and this facilitates various types of subsequent image processing.
FIGS. 7A-7E shows a process of conducting rotation correction processing on an enhanced image of a BD test card.
As for the BD test card, a blue channel is more advantageous to distinguish a stripe region from background information of the BD test card, and therefore, a blue-channel image is first extracted for the enhanced image of the BD test card ( FIG. 7A). According to the embodiment of the present disclosure, the blue-channel image thereof may be extracted after the size of the image may be reduced (e.g., reduced to 1/64 of the original image size), to reduce the overhead of calculation.
After the blue-channel image of the enhanced image is obtained, a fixed threshold 128 is used to conduct a binarization processing, to obtain a black-and-white binary image ( FIG. 7B). For example, in a case of 8-bit data, pixels of which blue component values are less than 128 are defined as white pixels; otherwise, the pixels are defined as black pixels.
A morphological operation is used for the black-and-white binary image of the BD test card to remove noises, and stripe regions are joined into a complete connected domain ( FIG. 7C). A connected domain analysis is conducted to find a maximum connected domain ( FIG. 7D), and at the same time, a rotation angle Θ is estimated according to an upper boundary of the connected domain. Selection correction is conducted based on the rotation angle Θ, to obtain a corrected result ( FIG. 7E). Moreover, rotation correction is conducted on the enhanced image of the BD test card based on the rotation angle Θ, to obtain a rotated enhanced image of the BD test card. According to the embodiment of the present disclosure, the above process may be repeated multiple times, to obtain a precise positioning result of the BD test card. FIG. 8A shows a process of conducting image segmentation processing on the rotated enhanced image of a BD test card, and FIG. 8B schematically shows a region of interest (ROI) of the BD test card after image segmentation processing.
As shown in FIG. 8A, according to the embodiment of the present disclosure, the step of conducting image segmentation on the rotated enhanced image of the BD test card includes:
recording the quantity of foreground pixels in each row of the blue-channel image after rotation as horizontal projection information (a curve part at the right side in FIG. 8A), and recording the quantity of foreground pixels in each column of the blue-channel image after rotation as vertical projection information (a curve part at the upper part in FIG. 8A); traversing the horizontal projection information in a direction from the center to the boundary, to obtain an upper boundary and a lower boundary of the ROI (as shown by two horizontal dashed lines in FIG. 8A), and traversing the vertical projection information in a direction from the center to the boundary, to obtain a left boundary and a right boundary of the ROI (as shown by two vertical dashed lines in FIG. 8A); and segmenting the rotated enhanced image of the BD test card according to the upper boundary, the lower boundary, the left boundary, and the right boundary, to obtain the ROI of the BD test card (as shown in FIG. 8B).
Here, the "foreground pixel" described in this text is such a pixel that in a certain region, the brightness of the foreground pixel is less than the average brightness of an image in this region. A reactant after disinfection processing assume apparent dark brown to black, and the brightness thereof is less than the brightness of other parts of the image data. Particularly, image enhancement processing is conducted on the image data such that the brightness distribution thereof is extended to the whole threshold range. Moreover, the blue-channel image of the enhanced image is extracted, and therefore, dark stripe parts are obviously distinguished from other parts of the image when counting statistics is conducted for the foreground pixels. When the horizontal projection information and the vertical projection information formed according to counts of the foreground pixels are traversed (in directions from the center to both sides), boundaries of the ROI may be determined by simply applying a threshold method. For example, when a difference between two adjacent count values of the foreground pixels is greater than a predetermined threshold, boundaries of the ROI may be determined.
A method for segmenting the rotated enhanced image is exemplified above with reference to FIG. 8A, but the present disclosure is not limited thereto. According to another embodiment of the present disclosure, the step of conducting image segmentation on the rotated enhanced image may include: obtaining a blue-channel image of the rotated enhanced image; binarizing the blue-channel image of the rotated enhanced image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain the ROI. The image processing process is similar to the image processing process described with reference to FIGS. 7A-7E, and therefore, repeated descriptions thereof are omitted herein. The above image processing is conducted based on the rotated image, and therefore, it is unnecessary to determine a rotation angle by using the obtained connected domain. Instead, boundaries of the ROI may be determined by using the obtained connected domain, and image segmentation is conducted on the rotated enhanced image by using the boundaries of the ROI, to obtain the ROI.
Referring to FIG. 5 again, at least one kind of single channel image data of the digital image (as shown in FIG. 8B) of the chemical test card may be acquired, and feature information may be extracted based on the at least one kind of single channel image data (S3). FIGS. 9A-9B shows a gray-channel image of an image in a ROI of a BD test card and a mask image thereof.
According to an embodiment of the present disclosure, the step of extracting feature information may include: obtaining a gray-channel image of an image in the ROI (as shown in FIG. 9A); calculating a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the ROI; segmenting the gray-channel image into a central region and a boundary region surrounding the central region, and calculating a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segmenting the gray-channel image into multiple blocks, and calculating a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtaining a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtaining a minimum block gray average from block gray averages corresponding to blocks in the boundary region. A four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average is the feature information.
In the above calculation process, the quantities of foreground pixels in different regions are used repeatedly. Therefore, a mask image may be constructed based on the ROI of the BD test card (as shown in FIG. 9B). In the mask image, the foreground pixels are displayed as white pixels (or marked as "1 "), and other parts are displayed as black pixels (or marked as "0"). A block segmentation manner of a gray-channel image and a segmentation manner of the central region are applied to the mask image, and the quantities (or count sums) of white pixels in each block, in the central region, and in the whole image are counted statistically respectively, so as to obtain the quantities of foreground pixels in each block, in the central region, and in the whole image. Moreover, the size (a dashed rectangular part) of a block for segmenting the gray-channel image is exemplified in FIG. 9A, but the present disclosure is not limited thereto, and the size of the block may be adjusted according to an actual application situation.
Moreover, according to the embodiment of the present disclosure, an overlapping portion may exist between adjacent blocks among the multiple blocks. Such a block segmentation manner can ensure that the calculated gray average has a desirable partial characteristic, and at the same time is more stable than the gray value of a single pixel. The segmented central region is exemplified in FIG. 9B, but the present disclosure is not limited thereto, and the size of the central region may be adjusted according to an actual application situation. It should be realized that, the same block segmentation manner and central region segmentation manner are applied to the gray -channel image and the mask image, and therefore, the block segmentation manner shown in the gray -channel image will be applied to the mask image, and the central region segmentation manner shown in the mask image will be applied to the gray-channel image.
In the four extracted gray averages, the maximum block gray average is generally extracted from a block located at the central region, and the minimum gray average is generally extracted from a block located at the boundary region. This is because that in the process of disinfection, high-temperature steam permeates from edges of the BD test card to the center, and a reactant located at the edge portion will change its color in the first place, and the color thereof is the most fully changed. The reactant may change its color from dark brown to black, and therefore, the reactant having the color fully changed may have a relatively low gray value. The minimum gray average acquired from the boundary region may be considered as the part of the BD test card where the reaction is the most fully conducted, and the maximum gray average acquired from the central region may be considered as the part of the BD test card where the reaction is the least fully conducted. The maximum gray average and the minimum gray average introduced into the feature information may reflect the uniformity of color change of the BD test card.
Moreover, the calculation of the gray average of each part depends on the quantity of foreground pixels of the corresponding part. As described above, the foreground pixels correspond to the dark stripe part of the BD test card (that is, a reactant region on the BD test card). In other words, when the gray average of each part is calculated, merely gray values of pixels corresponding to the reactant rather than pixels corresponding to the blank region (that is, a white stripe part) are considered.
The example of the extracting feature information based on the gray-channel image data of the digital image of the chemical test card is illustrated in the above embodiment; however, the present disclosure is not limited thereto. According to another embodiment of the present disclosure, the feature information may be extracted based on gray-channel image data and yellow-channel image data of the digital image of the chemical test card. For example, a full-image gray average and a central gray average may be calculated based on the yellow-channel image data, and the maximum block gray average and the minimum block gray average are calculated based on the gray-channel image.
Referring to FIG. 5 again, it may be judged according to the feature information whether a test conducted by using the chemical test card is successful (S4).
Specifically, weighted calculation may be conducted on the extracted feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, it may be determined that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, it may be determined that the test conducted by using the chemical test card is failed. Moreover, when the score is less than the first threshold and greater than or equal to the second threshold, it may be determined that the test conducted by using the chemical test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion. The expert includes (but is not limited to) a person who is very experienced in judging the BD test card, for example, a head nurse in a medical institution. Moreover, various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training. The acquisition of samples also depends on introduction of expert opinions.
According to the embodiment of the present disclosure, test samples in different cases may be obtained in advance according to judgment results of experienced experts, so as to conduct training to obtain various weights in the weighted calculation as well as the first threshold and the second threshold that are used for determining the extracted feature information. Moreover, in the process of judging the BD test card in an actual application, the expert opinion may be further introduced to a test result in a critical state (that is, between the first threshold and the second threshold), and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion, to constantly perfect the whole automatic reading method, such that the result of automatic reading may be better consistent with the expert opinion.
Modified exemplary embodiment: BD test card
FIG. 10 shows another example of image data of a BD test card.
As shown in FIG. 10, the image data of the BD test card includes a stripe region of the BD test card and an identification information region near the stripe region. In the example of FIG. 10, the identification information region is shown as a QR code. Moreover, only the upper half part of the image data of the BD test card is shown in the example shown in FIG. 10, and the whole image data further includes a lower part having an area almost the same as that of the shown part. As for the example of the image data of the BD test card shown in FIG. 10, the present disclosure provides a process of modified image processing.
According to an embodiment of the present disclosure, the process of modified image processing may include : obtaining a blue-channel image of the image data of the BD test card and reducing the image appropriately; binarizing the obtained image to obtain a black-and-white binary image; conducting a morphological operation on the black-and-white binary image and joining into a complete connected domain; and conducting connected domain analysis to obtain a maximum connected domain. The maximum connected domain may cover the stripe region of the BD test card. The process of modified image processing may further include: amplifying boundaries of the maximum connected domain proportionally, to cover an identification information region; and conducting image segmentation on the image data of the BD test card according to the amplified boundaries, to obtain a region of interest. According to the current embodiment, the region of interest not only includes the stripe region of the BD test card, but also includes the identification information region. Moreover, the process of modified image processing may further include: estimating a rotation angle of the image according to the boundaries of the maximum connection domain obtained previously (before amplification); and conducting rotation correction on the region of interest including the stripe region of the BD test card and the identification information region by using the obtained rotation angle, to obtain a rotated region of interest.
According to the current embodiment, the rotation angle of the image is estimated according to the boundaries of the maximum connected domain, and boundaries of the region of interest are obtained according to the boundaries of the amplified maximum connected domain. Image segmentation is conducted on the image data of the BD test card by using the boundaries of the region of interest, and then rotation correction is conducted by using the rotation angle on the region of interest obtained after segmentation. As for image data having a large area, conducting an operation of image segmentation before rotation correction may effectively reduce the overhead of calculation amount.
According to the current embodiment, the rotated region of interest includes the stripe region of the BD test card and the identification information region. A digital image of the stripe region of the BD test card may be acquired by using the manner described in the above embodiment, and feature information is extracted. Moreover, identification information of the BD test card may further be obtained by using a digital image of the identification information region. The BD test card may be authenticated by using the identification information of the BD test card. According to the embodiment of the present disclosure, before the step of extracting the digital image of the chemical test card, acquiring identification information on an outer package of the chemical test card (that is, first identification information) may also be included. The BD test card may be authenticated by using the first identification information and identification information of the BD test card (that is, second identification information). Subsequent processing (e.g., extracting feature information) is conducted on the acquired image data when the first identification information matches the second identification information. Moreover, according to the embodiment of the present disclosure, no subsequent processing is conducted when the first identification information does not match the second identification information, and it is prompted to scan the chemical test card again. The first identification information and the second identification information may be different from each other but have a unique corresponding relationship. In this way, test cards are prevented from being mixed up.
Exemplary embodiment: multi-parameter chemical integrator
The step of extracting the digital image of the chemical test card from the image data is illustrated in detail in the following by using a multi-parameter chemical integrator as an example.
It can be seen from the above analysis that, conducting automatic reading on a
multi-parameter chemical integrator (as shown in FIG. 4) adhered to a test report and determining a test result may facilitate digital management of the test result. Different from the case of the BD test card, an area of image data occupied by the multi-parameter chemical integrator is relatively small, and therefore, an operation of image segmentation may be conducted first before rotation correction in the process of extracting the digital image of the chemical test card, to reduce the overhead of the calculation amount.
Specifically, according to the embodiment of the present disclosure, when the chemical test card is the multi-parameter chemical integrator, the step of extracting the digital image of the chemical test card (S2) shown in FIG. 5 may include: conducting image enhancement processing on the acquired image data to obtain an enhanced image; conducting image segmentation on the enhanced image to obtain a ROI including the digital image of the chemical test card; and conducting rotation correction on an image in the acquired ROI.
FIG. 11 shows a diagram after image enhancement processing is conducted on the image data of the multi-parameter chemical integrator shown in FIG. 4; FIG. 12 shows a diagram after image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator shown in FIG. 11 ; FIG. 13 shows a diagram after rotation correction is conducted on a ROI of the multi-parameter chemical integrator shown in FIG. 12; FIG. 14 shows a diagram including an "ACCEPT" window obtained based on the ROI after rotation correction shown in FIG. 13; FIG. 15 shows a diagram of a black stripe including a multi-parameter chemical integrator after image segmentation and rotation correction are conducted; and FIG. 16 shows a diagram of a maximum connected domain including the black stripe obtained based on FIG. 15.
Referring to FIG. 4 and FIG. 11, according to the embodiment of the present disclosure, the step of conducting image enhancement processing on the image data of the multi-parameter chemical integrator may include: conducting color space conversion according to the following equation (2)
T = max(0, B-R) (2) wherein, T denotes a converted color space, and B denotes a blue-channel image of image data of the multi -parameter chemical integrator, and R denotes a red-channel image of image data of the multi -parameter chemical integrator.
In an actual application, the multi-parameter chemical integrator has a white bottom printed with a turquoise pattern. Specifically, an "ACCEPT" window of a PCD card is printed as a turquoise pattern having a large area. Therefore, the enhanced image after the image enhancement processing is conducive to subsequent determination of the position of the "ACCEPT" window.
According to the embodiment of the present disclosure, a rotation angle applied in the step of conducting rotation correction and boundaries applied in the step of conducting image segmentation may be obtained according to the black-and-white binary image of the enhanced image (that is, the image shown in FIG. 11) of the multi -parameter chemical integrator.
Specifically, according to the embodiment of the present disclosure, the step of conducting image segmentation on the enhanced image of the multi-parameter chemical integrator may include: binarizing the enhanced image of the multi-parameter chemical integrator to obtain a first black-and-white binary image. In the enhanced image of the multi -parameter chemical integrator after the color space conversion according to the above equation (2), the "ACCEPT" window that originally assumes turquoise may have a large contrast compared with display content of other parts. Therefore, after the binarization processing (that is, pixels of which brightness is greater than a threshold are defined as white pixels, and otherwise, pixels are defined as black pixels), the "ACCEPT" window displayed as white pixels may be clearly embodied in the obtained first black-and-white binary image. Then, a first maximum connected domain including the "ACCEPT" window may be obtained through connected domain analysis. The position of the "ACCEPT" window on the multi-parameter chemical integrator is relatively fixed, and therefore, boundaries of the digital image including the multi-parameter chemical integrator may be estimated by using the obtained first maximum connected domain, and image segmentation is conducted on the enhanced image of the multi-parameter chemical integrator according to the estimated boundaries to obtain a ROI. Moreover, a rotation angle may be estimated according to an upper boundary of the first maximum connected domain, and rotation correction may be conducted on the image (as shown in FIG. 12) in the ROI of the multi-parameter chemical integrator based on the rotation angle.
According to the embodiment of the present disclosure, when the chemical test card is the multi-parameter chemical integrator, the step of extracting feature information (S3) shown in FIG. 5 may include: binarizing the rotated image (as shown in FIG. 13) in the ROI to obtain a second black-and-white binary image; and conducting connected domain analysis on the second black-and-white binary image, to obtain a second maximum connected domain (as shown by the dashed block in FIG. 14). Different from the above first black-and-white binary image and the first maximum connected domain, the second black-and-white binary image and the second maximum connected domain based on the second black-and-white binary image are acquired from the image after segmentation and rotation. Therefore, the "ACCEPT" window included in the second maximum connected domain may have a basically vertical left boundary, thus facilitating subsequent judgment on whether a black stripe on the
multi-parameter chemical integrator "crawls" to the side of the "ACCEPT" window.
Moreover, the step of extracting feature information (S3) shown in FIG. 5 may further include a step of extracting a third maximum connected domain including the black stripe on the multi-parameter chemical integrator. Specifically, the step may include: acquiring a blue-channel image of the image data of the multi-parameter chemical integrator, and applying estimation of boundaries and a rotation angle according to the first maximum connected domain to the blue-channel image, to obtain a corrected feature image (as shown in FIG. 15); binarizing the corrected feature image to obtain a third black-and-white binary image; and conducting connected domain analysis on the third black-and-white binary image to obtain a third maximum connected domain (as shown by the dashed block in FIG. 16). Similar to the above second black-and-white binary image and the second maximum connected domain, the third black-and-white binary image and the third maximum connected domain based on the third black-and-white binary image are acquired from the image after segmentation and rotation. Therefore, the black stripe included in the third maximum connected domain may have a basically horizontal shape.
According to the embodiment of the present disclosure, the obtained second maximum connected domain and the third maximum connected domain are feature information extracted in the step of extracting feature information (S3) shown in the drawing. Subsequently, it may be judged according to the extracted feature information whether a test conducted by using the multi-parameter chemical integrator is successful (S4). Specifically, according to the embodiment of the present disclosure, when it is determined according to the extracted feature that the second maximum connected domain and the third maximum connected domain have an overlapping portion, it may be determined that the test conducted by using the multi-parameter chemical integrator is successful (that is, whether the black stripe "crawls" to the side of the "ACCEPT" window); otherwise, it may be determined that the test conducted by using the multi-parameter chemical integrator is failed.
The above embodiment illustrates the case in which the chemical test card is the multi-parameter chemical integrator by using a "crawling card" as an example, but the present disclosure is not limited thereto. It should be realized that the chemical test card automatic reading method according to the present disclosure may be applied to various multi-parameter chemical integrators used in various cases such as steam sterilization, ethylene oxide sterilization, and hydrogen peroxide sterilization.
Exemplary embodiment: automatic reading system
FIG. 17 schematically shows a block diagram of a chemical test card automatic reading system according to an exemplary embodiment.
According to another aspect of the present disclosure, a chemical test card automatic reading system is provided, to execute the chemical test card automatic reading method according to the present disclosure. As shown in FIG. 17, an automatic reading system 1000 according to an embodiment of the present disclosure may include an image acquiring module 1001, an image processing module 1002, a feature information extracting module 1003, and a test result judging module 1004. The image acquiring module 1001 is configured to acquire image data including a digital image of a chemical test card. The image processing module 1002 is configured to extract the digital image of the chemical test card from the acquired image data. The feature information extracting module 1003 is configured to acquire at least one kind of single channel image data of the digital image of the chemical test card extracted by the image processing module 1002, and extract feature information based on the at least one kind of single channel image data. The test result judging module 1004 is configured to judge according to the feature information extracted by the feature information extracting module 1003 whether a test conducted by using the chemical test card is successful.
According to the embodiment of the present disclosure, the image processing module 1002 may be configured to conduct image enhancement processing, rotation correction, and image segmentation on the image data acquired by the image acquiring module 1001, to extract the digital image of the chemical test card.
When the chemical test card is a BD test card, the image processing module 1002 may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct rotation correction on the enhanced image to obtain a rotated enhanced image; and conduct image segmentation on the rotated enhanced image to obtain a region of interest including the digital image of the chemical test card. The feature information extracting module 1003 may be configured to: obtain a gray-channel image of an image in the region of interest; calculate a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest; segment the gray-channel image into a central region and a boundary region surrounding the central region, and calculate a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segment the gray-channel image into multiple blocks, and calculate a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and obtain a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtain a minimum block gray average from block gray averages corresponding to blocks in the boundary region. A four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average may be the feature information. According to the embodiment of the present disclosure, an overlapping portion may exist between adjacent blocks among the multiple blocks.
The image enhancement processing, rotation correction and image segmentation conducted on the image data of the BD test card as well as the feature information extraction from the image in the region of interest are described with reference to steps S2 and S3 in FIG. 5 and with reference to FIG. 6 to FIGS. 9A-9B; therefore, repeated descriptions thereof are omitted herein.
The test result judging module 1004 may conduct weighted calculation on the feature information to obtain a score for the chemical test card. When the score is greater than or equal to a first threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is successful, and when the score is less than a second threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is failed. Various weights in the weighted calculation as well as the first threshold and the second threshold may be obtained by sample training. When the score is less than the first threshold and greater than or equal to the second threshold, the test result judging module 1004 may determine that the test conducted by using the BD test card is in a critical state. For the test in the critical state, an expert opinion may be introduced, and the first threshold and/or second threshold may be adjusted dynamically based on the expert opinion.
The judging according to the extracted feature information whether the test conducted by using the BD test card is successful has been described with reference to step S4 in FIG. 5, and therefore, repeated description thereof is omitted herein.
When the chemical test card is a multi-parameter chemical integrator, the image processing module 1002 may be configured to: conduct image enhancement processing on the acquired image data to obtain an enhanced image; conduct image segmentation on the enhanced image to obtain a region of interest including the digital image of the chemical test card; and conduct rotation correction on an image in the acquired region of interest.
The image enhancement processing, image segmentation, and rotation correction conducted on the image data of the multi-parameter chemical integrator as well as the feature information extraction from the image in the region of interest are described with reference to steps S2 and S3 in FIG. 5 and with reference to FIG. 11 to FIG. 16; therefore, repeated descriptions thereof are omitted herein.
According to the embodiment of the present disclosure, the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 may be implemented as program modules running on a computer, or may be implemented as dedicated (or universal) hardware modules that can execute corresponding functions.
Moreover, the image acquiring module 1001, the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 are shown as modules independent of each other; however, according to the embodiment of the present disclosure, these modules 1001 to 1004 may be all or partially implemented by using a single module. For example, a dedicated device having a scanning function may be implemented to implement all or partial functions of the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004.
According to the embodiment of the present disclosure, there may be such a module capable of implementing a part of function in each of the above modules. For example, the function of the image processing module 1002 may be specifically implemented by multiple sub-modules.
According to the embodiment of the present disclosure, the automatic reading system 1000 may further include a storing module configured to store, for example, software program codes for implementing the image processing module 1002, the feature information extracting module 1003, and the test result judging module 1004 and/or the first threshold and the second threshold for judging the score for the chemical test card.
According to the embodiment of the present disclosure, the automatic reading system 1000 may further include a displaying module configured to display, for example, a judgment result output by the test result judging module 1004 and/or display, for example, prompt information that an expert opinion needs to be provided when the test is in a critical state.
According to the embodiment of the present disclosure, the automatic reading system 1000 may be implemented as a system including (but not limited to) a scanner and a desktop computer. However, the present disclosure is not limited thereto, and according to another embodiment of the present disclosure, the automatic reading system 1000 may be implemented as a system including (but not limited to) a scanner and a portable mobile device, or the automatic reading system 1000 may be implemented as a dedicated device having a scanning function. Moreover, the automatic reading system 1000 may further include one or more modules and/or devices configured to implement various other additional functions.
According to the embodiment of the present disclosure, the automatic reading system 1000 may further include an identification information acquiring module 1005 and an identification information judging module 1006. The identification information acquiring module 1005 is configured to acquire first identification information on an outer package of the chemical test card. For example, the identification information acquiring module 1005 may be implemented as a bar code reading device, configured to read bar code or QR code information printed on the outer package of the chemical test card, to acquire the first identification information.
Moreover, before the feature information extracting module 1003 extracts feature information from the digital image of the chemical test card, the image processing module 1002 may first extract second identification information from the digital data of the chemical test card. The second identification information may be implemented as a bar code or a QR code (referring to an example shown in FIG. 10) printed on the chemical test card (or printed on a test report including the chemical test card). After the first identification information and the second identification information are extracted, the identification information judging module 1006 judges the first identification information and the second identification information. A subsequent processing (for example, extracting feature information by using the feature information extracting module 1003) may be conducted when the first identification information matches the second identification information. No subsequent processing is conducted when the first identification information does not match the second identification information, and it is prompted to scan the chemical test card again. The first identification information and the second identification information that correspond to the same chemical test card are different from each other, and they have a unique corresponding relationship. In this way, test cards are prevented from being mixed up.
Various advantages and effects of the exemplary embodiments are not limited to the above descriptions, and these advantages and functions may be easily understood through explanation of the specific embodiments in the present disclosure. The exemplary embodiments are shown and described in the above text; however, it will be apparent to those skilled in the art that modifications and changes may be made without departing from the scope of the present disclosure defined by claims.

Claims

1. A chemical test card automatic reading method, comprising:
acquiring image data comprising a digital image of a chemical test card;
extracting the digital image of the chemical test card from the acquired image data; acquiring at least one kind of single channel image data of the digital image of the chemical test card, and extracting feature information based on the at least one kind of single channel image data; and
judging according to the feature information whether a test conducted by using the chemical test card is successful.
2. The chemical test card automatic reading method according to claim 1, wherein the step of extracting the digital image of the chemical test card comprises conducting at least one of the following processing on the acquired image data: image enhancement processing, rotation correction, and image segmentation.
3. The chemical test card automatic reading method according to claim 1, wherein the chemical test card is a BD test card, and the step of extracting the digital image of the chemical test card comprises:
conducting image enhancement processing on the acquired image data to obtain an enhanced image;
conducting rotation correction on the enhanced image to obtain a rotated enhanced image; and
conducting image segmentation on the rotated enhanced image to obtain a region of interest comprising the digital image of the chemical test card.
4. The chemical test card automatic reading method according to claim 3, wherein the step of conducting image enhancement processing comprises:
converting the image data from a RGB color space to a YUV color space;
rding to the following equation (1)
Figure imgf000026_0001
(1) wherein, Y denotes a pixel value of the Y channel before adjustment, minY denotes a minimum value among pixel values of the Y channel before adjustment, maxY denotes a maximum value among pixel values of the Y channel before adjustment, Ymax denotes a maximum upper limit of the pixel value of the Y channel according to a digital level, and Y' denotes a pixel value of the Y channel after adjustment; and
converting the image data from the YUV color space to the RGB color space based on the pixel value of the Y channel after adjustment.
5. A chemical test card automatic reading system, comprising:
an image acquiring module, configured to acquire image data comprising a digital image of a chemical test card;
an image processing module, configured to extract the digital image of the chemical test card from the acquired image data;
a feature information extracting module, configured to acquire at least one kind of single channel image data of the digital image of the chemical test card extracted by the image processing module, and extract feature information based on the at least one kind of single channel image data; and
a test result judging module, configured to judge according to the feature information extracted by the feature information extracting module whether a test conducted by using the chemical test card is successful.
6. The chemical test card automatic reading system according to claim 5, wherein the image processing module is configured to conduct at least one of the following processing on the image data acquired by the image acquiring module: image enhancement processing, rotation correction, and image segmentation.
7. The chemical test card automatic reading system according to claim 5, wherein the chemical test card is a BD test card, and the image processing module is configured to:
conduct image enhancement processing on the acquired image data to obtain an enhanced image;
conduct rotation correction on the enhanced image to obtain a rotated enhanced image; and conduct image segmentation on the rotated enhanced image to obtain a region of interest comprising the digital image of the chemical test card.
8. The chemical test card automatic reading system according to claim 5, wherein the feature information extracting module is configured to:
obtain a gray-channel image of an image in the region of interest;
calculate a full-image gray average corresponding to the gray-channel image based on the quantity of foreground pixels in the region of interest, wherein the foreground pixel is a pixel whose brightness is less than an average brightness of the image in the region of interest;
segment the gray-channel image into a central region and a boundary region surrounding the central region, and calculate a central gray average corresponding to an image in the central region based on the quantity of foreground pixels in the central region; segment the gray -channel image into multiple blocks, and calculate a block gray average corresponding to an image in each block based on the quantity of foreground pixels in the block; and
obtain a maximum block gray average from block gray averages corresponding to blocks in the central region, and obtain a minimum block gray average from block gray averages corresponding to blocks in the boundary region,
wherein, a four-dimensional feature vector consisting of the full-image gray average, the central gray average, the maximum block gray average, and the minimum block gray average is the feature information.
9. The chemical test card automatic reading system according to claim 8, wherein adjacent blocks in the multiple blocks overlap each other.
10. The chemical test card automatic reading system according to claim 8, wherein the test result judging module conducts weighted calculation on the feature information to obtain a score for the chemical test card, and
when the score is greater than or equal to a first threshold, the test result judging module determines that the test conducted by using the chemical test card is successful, and when the score is less than a second threshold, the test result judging module determines that the test conducted by using the chemical test card is failed.
11. The chemical test card automatic reading system according to claim 10, wherein various weights in the weighted calculation as well as the first threshold and the second threshold are obtained by sample training.
12. The chemical test card automatic reading system according to claim 5, wherein the image acquiring module is a scanner.
13. The chemical test card automatic reading system according to claim 5, wherein the feature information extracting module is configured to acquire only one kind of single channel image data of the digital image of the chemical test card, and extract feature information based on the kind of single channel image data.
14. The chemical test card automatic reading system according to claim 5, wherein the chemical test card is a multi-parameter chemical integrator, and the image processing module is configured to:
conduct image enhancement processing on the acquired image data to obtain an enhanced image;
conduct image segmentation on the enhanced image to obtain a region of interest comprising the digital image of the chemical test card; and
conduct rotation correction on an image in the acquired region of interest.
15. The chemical test card automatic reading system according to claim 5, further comprising an identification information acquiring module and an identification information judging module, the identification information acquiring module being configured to acquire first identification information on an outer package of the chemical test card,
wherein the image acquiring module further acquires second identification information on the chemical test card; and
the identification information judging module is configured to judge the first identification information and the second identification information, and the automatic reading system conducts subsequent processing on the image data acquired by the image acquiring module when the first identification information matches the second identification information.
PCT/US2017/038466 2016-06-23 2017-06-21 Chemical test card automatic reading method and system WO2017223159A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610463801.0 2016-06-23
CN201610463801.0A CN106127212B (en) 2016-06-23 2016-06-23 The automatic read method and system of test chemical card

Publications (1)

Publication Number Publication Date
WO2017223159A1 true WO2017223159A1 (en) 2017-12-28

Family

ID=57268095

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2017/038466 WO2017223159A1 (en) 2016-06-23 2017-06-21 Chemical test card automatic reading method and system

Country Status (2)

Country Link
CN (1) CN106127212B (en)
WO (1) WO2017223159A1 (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022018419A1 (en) * 2020-07-20 2022-01-27 Tristel Plc Validation of mixing procedures in a decontamination process
CN114252284A (en) * 2021-12-07 2022-03-29 聊城鲁西双氧水新材料科技有限公司 Judgment method for trial production conditions of hydrogen peroxide production device

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108181328A (en) * 2017-12-08 2018-06-19 中国科学院重庆绿色智能技术研究院 A kind of method and system for being used to detect X-ray machine material discrimination power
CN108846312B (en) * 2018-05-07 2021-03-12 深圳市恒扬数据股份有限公司 Identification method and device for effective activity area of bacteria and terminal equipment
CN113343988A (en) * 2021-07-06 2021-09-03 北矿机电科技有限责任公司 Method, system and device for identifying ore belt of shaking table

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103971358A (en) * 2013-02-05 2014-08-06 深圳市深图医学影像设备有限公司 Image interest area display optimizing method and device system

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
ADRIAN CARRIO ET AL: "Automated Low-Cost Smartphone-Based Lateral Flow Saliva Test Reader for Drugs-of-Abuse Detection", SENSORS, vol. 15, no. 12, 24 November 2015 (2015-11-24), pages 29569 - 29593, XP055257628, DOI: 10.3390/s151129569 *
LUIS FERMÍN CAPITÁN-VALLVEY ET AL: "Recent developments in computer vision-based analytical chemistry: A tutorial review", ANALYTICA CHIMICA ACTA, vol. 899, 1 October 2015 (2015-10-01), AMSTERDAM, NL, pages 23 - 56, XP055400184, ISSN: 0003-2670, DOI: 10.1016/j.aca.2015.10.009 *
NURIA LOPEZ-RUIZ ET AL: "Smartphone-Based Simultaneous pH and Nitrite Colorimetric Determination for Paper Microfluidic Devices", ANALYTICAL CHEMISTRY, vol. 86, no. 19, 7 October 2014 (2014-10-07), US, pages 9554 - 9562, XP055241637, ISSN: 0003-2700, DOI: 10.1021/ac5019205 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022018419A1 (en) * 2020-07-20 2022-01-27 Tristel Plc Validation of mixing procedures in a decontamination process
CN114252284A (en) * 2021-12-07 2022-03-29 聊城鲁西双氧水新材料科技有限公司 Judgment method for trial production conditions of hydrogen peroxide production device

Also Published As

Publication number Publication date
CN106127212A (en) 2016-11-16
CN106127212B (en) 2019-06-25

Similar Documents

Publication Publication Date Title
WO2017223159A1 (en) Chemical test card automatic reading method and system
US8184900B2 (en) Automatic detection and correction of non-red eye flash defects
US7830418B2 (en) Perceptually-derived red-eye correction
US9721532B2 (en) Color chart detection apparatus, color chart detection method, and color chart detection computer program
US20160188937A1 (en) Universal assay reader
US9171224B2 (en) Method of improving contrast for text extraction and recognition applications
US8514240B2 (en) Authentication system, authentication program, and method of authentication
CN108389215B (en) Edge detection method and device, computer storage medium and terminal
CN110415212A (en) Abnormal cell detection method, device and computer readable storage medium
JP6630341B2 (en) Optical detection of symbols
CN106934411A (en) Electronic paper marking method based on template matches
JP2008187710A (en) Method and system for specifying image characteristics, method for determining background color, control program and recording medium
CN113569859B (en) Image processing method and device, electronic equipment and storage medium
JP2009038737A (en) Image processing apparatus
JP6583316B2 (en) Identification device, identification method, and program
CN111445402A (en) Image denoising method and device
Datta Credit Card Processing Using Cell Phone Images
CN114037664A (en) Infusion set accessory detection method and system based on machine vision, storage medium and equipment
WO2017128273A1 (en) Error block determination
Cudek et al. Melanocytic skin lesion image classification. Part I: Recognition of skin lesion
CN111046899A (en) Method, device and equipment for identifying authenticity of identity card and storage medium
CN103679684A (en) Device, method and electronic equipment for detecting cloud in image
CN115049746B (en) Image color correction method, device, electronic equipment and storage medium
US20240169529A1 (en) Validating a sample collection device
US20220375096A1 (en) System and method to detect quality of silk cocoons

Legal Events

Date Code Title Description
DPE1 Request for preliminary examination filed after expiration of 19th month from priority date (pct application filed from 20040101)
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17736818

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 17736818

Country of ref document: EP

Kind code of ref document: A1