Intensity disparity compensation in images for colony counting

Abstract

Embodiments described herein relate to determining a region of interest on a growth medium. A total brightness control value of a plurality of illumination sources configured to illuminate the growth medium is calculated. The total brightness control value generates at least one image that substantially matches a target intensity at the region of interest. A separate brightness value for each illumination source of the plurality of illumination sources is calculated by individually adjusting the brightness of each illumination source to generate at least one image that substantially matches the target intensity in an area of influence of each respective illumination source. A calibrated brightness value for each illumination source is determined based on the image intensity when each illumination source is turned on at the respective separate brightness value and the intensity generated within each respective area of influence when each illumination source is turned on individually.

Description

Technical Field

[0001] This disclosure relates throughout to imaging apparatus. More specifically, this disclosure relates to compensating for differences in image intensity. Background Technology

[0002] For food product developers and distributors, testing for biological contamination in food or other materials has become important and often mandatory. Biological testing is also used to identify bacteria or other agents in samples such as laboratory samples (e.g., blood samples from medical patients), laboratory samples developed for experimental purposes, and other types of biological samples. Various technologies and devices can be used to improve biological testing and to streamline and standardize the testing process.

[0003] Biological growth plates can be used to count or identify the presence of bacteria, thereby enabling improved assays (in the case of food testing) or accurate diagnoses (in the case of medical applications). In other applications, biological growth plates can be used to rapidly grow bacteria or other biological agents in laboratory samples, for example, for experimental purposes.

[0004] Several biological growth plates have been developed. As an example, 3M Company of St. Paul, Minn (hereinafter referred to as "3M") has developed a biological growth plate. This biological growth plate is sold by 3M under the trade name PETRIFILM. Biological growth plates can be used to facilitate the rapid growth and detection of bacteria or other biological agents often associated with food contamination, including, for example, aerobic bacteria, Escherichia coli, coliform bacteria, Enterobacteriaceae, yeast, molds, Staphylococcus aureus, Listeria, Campylobacter, etc. The use of PETRIFILM plates or other growth media simplifies bacterial testing of food samples. Summary of the Invention

[0005] The embodiments described herein relate to a method comprising determining a region of interest on a growth medium. A total brightness control value is calculated for a plurality of illumination sources configured to irradiate the growth medium. The total brightness control value generates at least one image substantially matching the target intensity at the region of interest. An individual brightness value for each of the plurality of illumination sources is calculated based on the total brightness value by individually adjusting the brightness of each illumination source to generate at least one image substantially matching the target intensity in the region of influence of each respective illumination source. A calibrated brightness value for each illumination source is determined based on the image intensity when each illumination source is turned on with its corresponding individual brightness value, and the intensity generated by each illumination source in each respective region of influence when turned on individually.

[0006] The implementation relates to a system including a processor and a memory storing computer program instructions that, when executed by the processor, cause the processor to perform operations. The operations include determining a region of interest on a growth medium. A total brightness control value is calculated for a plurality of illumination sources configured to irradiate the growth medium. The total brightness control value generates at least one image that substantially matches the target intensity at the region of interest. An individual brightness value for each of the plurality of illumination sources is calculated based on the total brightness value by individually adjusting the brightness of each illumination source to generate at least one image that substantially matches the target intensity in the region of influence of each corresponding illumination source. A calibrated brightness value for each illumination source is determined based on the image intensity when each illumination source is turned on with its corresponding individual brightness value, and the intensity generated by each illumination source in each corresponding region of influence when turned on individually.

[0007] The implementation relates to a non-transitory computer-readable medium storing computer program instructions for designing microstructures, which, when executed by a processor, cause the processor to perform operations. These operations include determining a region of interest on a growth medium. A total brightness control value is calculated for a plurality of illumination sources configured to irradiate the growth medium. The total brightness control value generates at least one image substantially matching the target intensity at the region of interest. An individual brightness value for each of the plurality of illumination sources is calculated based on the total brightness value by individually adjusting the brightness of each illumination source to generate at least one image substantially matching the target intensity in the region of influence of each corresponding illumination source. A calibrated brightness value for each illumination source is determined based on the image intensity when each illumination source is turned on with its corresponding individual brightness value, and the intensity generated by each illumination source in each corresponding region of influence when turned on individually.

[0008] The foregoing description is not intended to depict every embodiment or every specific implementation. Rather, a more complete understanding of the exemplary embodiments will become apparent from the accompanying drawings and the following detailed description of exemplary embodiments and claims. Attached Figure Description

[0009] The present technology can be more fully understood and appreciated in light of the following detailed description of various embodiments in conjunction with the accompanying drawings.

[0010] Figures 1A to 1C A growth medium with background material on the culture medium, according to an embodiment described herein, is shown, which interferes with the readings of the manual counter and the apparatus.

[0011] Figures 2A to 2D Different types of colony characteristics of species according to the embodiments described herein are shown;

[0012] Figure 3 Examples of different colony morphologies that may occur when the colony count is high and / or the colonies are closely packed together on the culture medium, according to the embodiments described herein;

[0013] Figure 4 The illustration shows relatively small and / or pale colonies compared to the background culture medium, according to the embodiments described herein;

[0014] Figure 5 An example of a diffused colony according to the embodiment described herein is shown;

[0015] Figure 6A An example of a non-grown plate according to the embodiment described herein is shown;

[0016] Figure 6B and Figure 6C An example view showing a growth medium brightening due to microbial growth according to the embodiments described herein;

[0017] Figure 7 The process for compensating for differences in optical output according to the embodiment described herein is illustrated;

[0018] Figure 8 A more detailed process for calculating the total brightness control value according to the embodiment described herein is shown;

[0019] Figure 9 The use of the implementation scheme described herein is illustrated. Figure 8 An example of a heatmap of the region of interest for which the total brightness control value is calculated;

[0020] Figure 10 A more detailed process for calculating individual luminance values ​​according to the embodiments described herein is shown;

[0021] Figures 11A to 11D Exemplary areas of interest are shown with different individual lighting sources turned on according to the embodiments described herein;

[0022] Figure 12 A more detailed process for calculating calibration luminance values ​​according to the embodiments described herein is shown;

[0023] Figure 13A An example of an image taken using the final calibrated brightness value according to the embodiment described herein is shown;

[0024] Figure 13B The embodiments described herein are shown. Figure 13A The heatmap of the image;

[0025] Figure 14 The process of aligning the spectrum according to the embodiment described herein is illustrated;

[0026] Figure 15A and Figure 15B Examples of misaligned and aligned normal histograms according to the embodiments described herein are shown respectively;

[0027] Figure 16A and Figure 16B Examples of misaligned nonnormal histograms and aligned nonnormal histograms according to the embodiments described herein are shown respectively;

[0028] Figure 17 The process for generating a normalization factor according to the implementation scheme described herein is illustrated;

[0029] Figure 18 An example is shown of how normalization can be used to produce a uniform image over a 5×5 sliding window according to the implementation described herein;

[0030] Figure 19A and Figure 19B An exemplary image is shown using only the calibration lighting system described herein;

[0031] Figure 20 An exemplary histogram according to the embodiment described herein is shown, which explains the light distribution of each red, green, and blue channel and the effect of flat-field normalization on image uniformity;

[0032] Figures 21A to 21F The following diagram shows heatmap representations of each of the red, green, and blue channels before and after normalization according to the embodiment described herein; and

[0033] Figure 22 A high-level block diagram of a system configured to implement the processes described herein is shown.

[0034] The accompanying drawings are primarily for clarity and therefore not necessarily drawn to scale. Furthermore, various structures / components, including but not limited to fasteners, electronic components (wires, cables, etc.), etc., may be shown schematically or removed from some or all views to better illustrate aspects of the depicted embodiments, or the inclusion of such structures / components may not be necessary for understanding the various exemplary embodiments described herein. However, the absence of illustrations / descriptions of such structures / components in certain figures should not be construed as limiting the scope of the various embodiments in any way. Detailed Implementation

[0035] The food industry routinely monitors levels of indicator groups of microorganisms. These organisms are considered quality and hygiene indicators and may pose health hazards to humans. A common method for quantifying indicator organisms is plate counting (colony counting). This is accomplished by culturing the target microorganism on suitable growth cultures such as dry film or agar plates and counting the colonies. Millions of colony counts have been performed in the food industry globally by food producers, reference laboratories, food processors, and manufacturers. Colony counting can be performed at multiple stages, including counting of feedstocks, processed products, finished products, and / or environmental samples taken from the food processing environment.

[0036] Counting colonies on growth media is highly complex and can greatly benefit from improvements in the quality of growth medium images. The unique challenges and requirements of this field complicate this work, examples of which include, but are not limited to, the challenges of interpreting growth media and the accuracy requirements of colony counting devices. These requirements necessitate the development of large amounts of data when developing robust systems. The diverse appearances and characteristics that can occur on growth media make colony interpretation challenging for both human operators and automated colony counting solutions.

[0037] Organisms cultured on growth media can be distinguished from a background of various samples. These samples may include one or more of agricultural products, dairy products, meat and poultry, confectionery, processed foods, raw materials, and / or environmental samples from food production facilities. Manual counters are typically used to scan or count the amount of bacterial colonies or specific biological agents on biological growth plates. Using imaging devices that automate the detection and / or counting of bacterial colonies and / or other biological agents can improve the accuracy, efficiency, and / or consistency of colony detection and / or counting. For example, food or laboratory samples can be placed on a biological growth plate, which can then be inserted into a culture chamber. After incubation, the biological growth plate can be placed in an imaging device for automated detection and counting of bacterial growth. This imaging device automates the detection and counting of bacteria or other biological agents on the biological growth plate, thus improving this biological testing process by reducing human error.

[0038] These samples deposited background materials (fibers, particles, colors) on the culture medium, which interfered with, for example... Figures 1A to 1C The manual counter and device interpretation shown in the example. Figure 1A An example of background material 110 on a culture medium is shown. Figure 1B An example of colony 120 without background material is shown. Figure 1C The growth medium shown is a mixture 130 of particles and colonies on a growth medium.

[0039] Colony characteristics are diverse, including size, shape, color, bubble production, and regional distribution. Some culture media are used to culture multiple types of organisms. In these cases, it is essential to identify colony characteristics to distinguish relevant colonies from other types. Figures 2A to 2D Different types of colony characteristics are shown. Figure 2A The image shows colony 210 producing bubbles on a growth medium. (Example:) Figure 2B As shown, a mixture of gas-producing colonies 240 and non-gas-producing colonies 230 may exist. Distinguishing between these different types of colonies can be useful. Depending on the configuration, irregular bubbles 250 that are not gas-related may be present on the growth medium, such as... Figure 2C As shown. Distinguishing between air bubbles and non-gas-related bubbles can be useful, so that non-gas-related bubbles are not included in the colony count. Depending on the configuration, colonies can generate acidic regions of varying intensities on the growth medium.260 Figure 2D As shown. It may be useful to distinguish between colonies that produce acidic regions and those that do not.

[0040] Figure 3 Examples of different colony morphologies that may occur, such as when colony counts are high and / or colonies are closely packed together on the culture medium, are shown. Identifying and / or distinguishing all colonies with different morphologies can be useful. When colonies are close together, it can be difficult to distinguish the different morphologies.

[0041] In some cases, colonies may be relatively small and / or pale compared to the background medium, such as 410. Figure 4 As shown. This can make it difficult to detect the presence of bacterial colonies.

[0042] In some configurations, colonies can spread, cover the growth medium, and / or alter its appearance. This type of growth can affect a region of the medium or the entire growth plate. These conditions also make growth difficult to detect. Figure 5 An example of spreading colonies is shown.

[0043] Figure 6A An example of a plate that has not grown is shown. Figure 6B and Figure 6C An example view showing the growth medium brightening due to microbial growth is shown. Figures 6A to 6C In such cases, it may be difficult to identify any bacterial growth from the growth medium area.

[0044] Figures 1A to 6C The examples described illustrate some of the complexities and challenges that colony counting devices need to overcome. The consequences of erroneous results can include product recalls, rejections, significant losses, damage to brand reputation, and / or harm to consumers.

[0045] Apparatus used to irradiate growth media to detect microbial growth can be affected by variations in the output of the light source. Compensating for these variations in the light source, light diffusion, light reflection, lens spatial efficiency, and / or camera response can contribute to a highly consistent response of light levels in the images from the colony counting apparatus. Images with consistent brightness allow for accurate bacterial colony identification and / or classification results, as each localized colony area can have a constant response to isolated colonies from food samples. The apparatus and methods described herein can be used to distinguish colonies from other contaminants in a sample, including pH differences and / or other sources that can cause changes in the culture medium background. The embodiments described herein involve compensating for variations in the output of the light source and the response of the image capture system to the light it produces. The image capture system may include multiple components in addition to the light source. For example, the image capture system may include one or more reflectors, lenses, cameras, and optical paths that may be affected by variations in the light source. Compensating for variations in the output of the light source allows for a uniform image response, enabling accurate detection of changes in the culture medium. The processes described herein may be performed once in a factory, for example, and / or can be repeated in the field. For example, the processes described herein may be repeated periodically and / or as needed by the user.

[0046] Figure 7 A process for compensating for differences in light output according to the embodiments described herein is illustrated. A region of interest 710 on the growth medium is determined. The region of interest may also be referred to herein as a region of interest (ROI). The region of interest may include any region on the growth medium containing the object of interest. For example, the region of interest may include a region on the growth medium containing bacterial growth. For example, the growth medium may include a biological growth plate.

[0047] Calculate a total brightness control value to generate at least one image that substantially matches the target intensity of 720 at the region of interest of multiple illumination sources. The target intensity can be a fixed value, chosen to ensure a sufficient intensity range above the target value so that a bright image still contains information useful for counting. The target value allows for images smaller than the saturation value and does not result in a large number of pixels having a maximum value of 255. The target intensity can be selected by evaluating a desired series of images and determining the range of desired maximum brightness.

[0048] This document uses the terms "brightness" and "intensity" in various references. It should be understood that "brightness" is generally used to refer to the light output of an illumination source, and "intensity" is used to refer to the brightness of an image captured using an illumination source. An illumination source can be any suitable lighting device. For example, illumination sources may include light-emitting diodes (LEDs), incandescent devices, bioluminescent devices, lasers, reflected sunlight, combustion devices (such as oil lamps or candles), and / or fluorescent devices. Depending on the configuration, calculating the total brightness control value involves simultaneously adjusting the brightness of all illumination sources in the spectrum to a brightness level equal to or below the target intensity.

[0049] Multiple images of the region of interest are captured iteratively, starting with a common brightness value that increases from below the target intensity. An average intensity can be captured for each image, and a total brightness control value can be calculated based on this average value.

[0050] The individual luminance value for each of a plurality of illumination sources is calculated by individually adjusting the luminance of each illumination source to generate at least one image that substantially matches the target intensity in the region of influence of each corresponding illumination source. 730. According to various configurations described herein, calculating the individual luminance value involves: for each illumination source in the spectrum, iteratively capturing multiple images of the region of interest for each illumination source at luminance levels increasing from a luminance level below a luminance control value. Depending on various configurations, images are captured when each illumination source is the only illuminated source. In some cases, multiple images are captured starting from a certain luminance level such that the intensity of the images is significantly lower than the target level, where one illumination source is activated at a time. The average intensity of each captured image in the region of influence of the activated illumination sources can be calculated, and the individual luminance value for each illumination source can be calculated based on this average value.

[0051] The calibrated luminance value 740 for each illumination source can be determined based on the total luminance control value and individual luminance values. According to the various embodiments described herein, determining the calibrated luminance value for each illumination source includes calculating the difference between the target intensity value and the individual intensity at each individual luminance level within the influence area of ​​the light source. According to the various embodiments, the calibrated luminance value for each illumination source is determined based on: the image intensity when each illumination source is turned on with its corresponding individual luminance value, and the intensity generated by each illumination source in each corresponding influence area when individually turned on.

[0052] According to the various embodiments described herein, a percentage change can be calculated to compensate for the difference in each individual luminance value. Multiple images of the region of interest are captured. An average intensity can be captured for each captured image, and a calibrated luminance value for each illumination source can be determined based on the average value and the percentage change value. Depending on the configuration, using the average intensity of the region of interest, the percentage difference between the average intensity of the region of interest and the target intensity is determined. The data captured for each individual light source can be used to determine the amount by which the luminance of each light source needs to be adjusted to move its intensity by the same percentage. The percentage by which the image intensity of each individual light source is changed is the same as the percentage of the captured intensity relative to the target intensity, which can be the individual light source luminance value.

[0053] Can be performed on one or more spectra Figure 7 The steps are shown. For example, they can be performed for three spectra such as red, green, and blue. Figure 7 The steps are as follows. Red, green, and blue represent primary colors, but other colors may be used in addition to primary colors or as alternatives to primary colors. For example, any color from the visible, infrared, and / or ultraviolet spectrum can be used.

[0054] Figure 8 A more detailed process for calculating the total brightness control value according to the embodiment described herein is shown. It should be understood that... Figure 8 The other flowcharts shown herein can be performed for a single spectrum or repeated for other spectra. The region of interest 810 where colony counting will be performed is determined. The region of interest can be determined by identifying the boundaries of a clear image region on the device image. More specifically, the region of interest for counting can be found by searching for known darker areas in the image representing the region boundaries. Dark areas can be defined by one or more boxes surrounding the region of interest that block light from the image plane, or by areas that are not illuminated so that backlighting only covers the counting area. The brightness of all illumination sources for each spectrum is calculated so that the average intensity of the region of interest substantially matches the target intensity 815. Simultaneously, all illumination sources are adjusted to brightness levels below the target intensity level, allowing individual adjustment of brightness levels 820. Depending on various configurations, all illumination sources are simultaneously adjusted to brightness levels that generate an intensity just below the target level.

[0055] Multiple images 830 are captured iteratively at increasing brightness levels. The intensity at the initial brightness level of each captured image can be significantly lower than the target intensity value without causing the intensity to reach a minimum value (e.g., an intensity of 0). The minimum value can represent the lowest black level observable by the image capture device. The brightness level can be adjusted low enough to ensure that the resulting image intensity is significantly lower than the target intensity level. This low initial value can be used to correctly create a linear equation representing the intensity response to any brightness. This equation is then used to calculate the brightness that generates the intensity that best matches the target value. This low starting point is created by analyzing the response and calibration conditions of the given image system and setting a value low enough below the target that any difference in the system will not cause the level to exceed the target intensity. In some cases, the initial brightness is adjusted to be about half or less of the target intensity. Depending on the configuration, small increments are used to increase the brightness level of each subsequent image. Depending on the configuration, the increments are small enough to provide enough points to create an accurate equation between the start and end. Capturing a larger number of images increases the required time. Therefore, it may be necessary to balance the required time and generate an accurate equation that increases the brightness level until an image intensity just below the saturation value is reached. For example, brightness can be increased until just before any pixel in the region of interest becomes white. Calculate the average intensity 840 for each of the multiple images. Use the brightness level and average intensity to create a best-fit equation 850. Any suitable technique can be used to create the best-fit equation. For example, one or more of linear regression, second-order curve fitting, multi-point averaging, and curve smoothing can be used to create the best-fit equation. Calculate the total brightness control value, which will essentially generate the target average intensity 860. Depending on the various configurations, the total brightness control value is calculated by solving the best-fit equation for the target intensity.

[0056] Determine if all total luminance control values ​​have been calculated for each spectrum 870. If it is determined at 870 that not all total luminance control values ​​have been calculated for all spectra, the process continues to calculate the luminance of the illumination source for another spectrum 815. If it is determined that all total luminance control values ​​have been calculated, the process completes 880, and the calculated total luminance control values ​​are used in subsequent steps.

[0057] Figure 9 It shows the use of Figure 8 An example of a heatmap of the region of interest for which the total brightness control value is calculated. As can be observed, most of the region of interest is relatively uniform, but some areas, particularly along the periphery, are darker than the rest of the region, as shown at 910. Figure 9 The heatmap shows uniformity. In other words, each pixel in the heatmap represents the offset of the intensity relative to the target intensity value. Figure 9In the diagram, most values ​​are offset by +1 or +2 relative to the target intensity value. Depending on the specific implementation, the target intensity value is approximately 200, so most heatmaps are close to or equal to 200. The darker areas shown in 910 represent negative offsets. For example, one or more pixel intensity values ​​in a darker area could be offset by -9 (200 - 9 = 191) or -14 (200 - 14 = 186).

[0058] Figure 10 A more detailed process for calculating individual luminance values ​​according to the embodiment described herein is shown. The influence area 1010 of the illumination source is determined. This is accomplished by capturing at least one image only when the illumination source is on and determining the region where the image intensity of the captured image is a threshold percentage of the target intensity. In some cases, the threshold percentage may range from approximately 85% to approximately 100%. Depending on various configurations, the threshold percentage is approximately 95%.

[0059] For the spectrum being tested, to range from below... Figure 8 The system iteratively captures multiple images 1030 of the area affected by the illumination source, starting with an increasing brightness level based on the calculated total brightness control value. Depending on various configurations, images of one light source are captured at a time. The brightness level can be applied to one light source at a time and / or cycled through the entire range of a light source. In some cases, images are captured starting at a brightness level significantly lower than the total brightness control value. For example, images can be captured starting at a brightness level that will give the image at least 1 / 2 of the target intensity without causing the intensity to reach 0. Depending on various configurations, the brightness level of each subsequent image is increased using small increments. Depending on various configurations, at least five images are captured at different brightness levels to create a well-representative equation. The brightness level of each captured image can be increased until an image intensity above the target intensity value but below the saturation value is reached. Depending on various configurations, the brightness level is increased to a level significantly higher than the total brightness control value. Depending on various configurations, the brightness level is increased until at least two points tested are above the target intensity. The brightness level can be increased until approximately 120% of the target intensity is reached without causing saturation. The average intensity 1040 of each image in the multiple images at the area affected by each illumination source is calculated. The average intensity can be calculated for each light source connected in series or parallel. A best-fit equation 1050 is created using the luminance level and average intensity of a single illumination source. Any suitable technique can be used to create the best-fit equation. For example, linear regression can be used. Individual luminance values ​​are calculated, which will generate the target average intensity 1060 within the influence area of ​​the single illumination source being tested. Depending on various configurations, the individual luminance values ​​are calculated by solving the best-fit equation for the target average intensity.

[0060] Determine if all individual illumination source luminance values ​​have been calculated for the current spectrum 1070. If it is determined at 1070 that not all illumination source luminance levels have been calculated, the process continues to calculate the luminance values ​​for another illumination source in the current spectrum being tested 1020. If it is determined that all individual illumination source luminance values ​​have been calculated, determine if all individual illumination source luminance values ​​have been calculated for all spectra 1080. If it is determined that not all illumination source luminance values ​​have been calculated for all spectra, the process continues to calculate the individual illumination source luminance values ​​for other spectra. If it is determined that all illumination source luminance values ​​have been calculated for all spectra, the process completes 1090, and the calculated individual luminance values ​​are used in subsequent steps of the calibration process.

[0061] Figures 11A to 11D Exemplary areas of interest are shown with different individual lighting sources on. As can be observed, each lighting source has a different area of ​​influence, as demonstrated by the brighter areas in different zones of the area of ​​interest.

[0062] Figure 12 A more detailed process for calculating calibrated luminance values ​​according to the embodiment described herein is shown. All lighting sources are... Figure 10 Their corresponding individual brightness values ​​are calculated and set to 1210.

[0063] Multiple images are iteratively captured at substantially the same luminance level determined for each illumination source to generate images within their influence regions that match the target intensity level of the spectrum being tested 1230. The luminance level used may be the same as or different from the luminance level of a different spectrum being tested. The average intensity of each of the multiple images is calculated 1240. The total average of all average intensities from all images is determined 1250. The amount of variation required to achieve the target intensity is calculated by subtracting the total average from the target intensity 1260. For each illumination source, a value will be determined that will give a... Figure 12 The calibrated luminance value is 1270, which is the same amount of intensity variation calculated for each individual luminance value.

[0064] Determine whether all calibrated luminance values ​​have been calculated for the current spectrum being tested 1282. If it is determined at 1282 that not all calibrated luminance values ​​have been calculated for the current spectrum, the process continues to capture more images of the current spectrum being tested 1230. If it is determined at 1282 that all calibrated luminance values ​​have been calculated for the current spectrum, determine whether all calibrated luminance values ​​have been calculated for all spectra 1284. If it is determined at 1284 that not all calibrated luminance values ​​have been calculated for all spectra, the process continues to turn on all illumination sources with individual luminance values ​​for the next spectrum to be tested 1290. If it is determined at 1284 that all calibrated luminance values ​​have been calculated for all spectra, the process completes 1290.

[0065] Figure 13A and Figure 13B An example of an image using the final calibrated brightness values ​​is shown. Figure 13A ) and heatmap of the image ( Figure 13B ). If possible Figure 13B The heatmap observed is largely uniform, except for some minor inhomogeneities along the perimeter 1310.

[0066] After the illumination setup is calibrated, the spectra can be further aligned to make the brightness response to intensity as similar as possible between spectra and between setups. For example, the intensity histogram of each spectrum can be used to further refine the results. Each spectrum can be aligned by slightly increasing or decreasing the brightness control of that spectrum across all its illumination sources.

[0067] Figure 14 The process of aligning the spectrum according to the embodiment described herein is illustrated. Multiple images 1410 of the region of interest are captured. Depending on various configurations, multiple images are captured without altering the brightness of the illumination source for the current spectrum being tested. The pixel intensity of each image is averaged to form an average pixel intensity 1420. A histogram of pixels for the captured current spectrum is generated for each intensity level 1430. The average intensity value and peak value of the average pixel intensity for each histogram are determined 1440. It is determined whether all spectral histograms have been generated 1450. If it is determined at 1450 that not all spectral histograms have been generated, multiple images are captured for another spectrum and the process continues. If it is determined at 1460 that all spectral histograms have been generated, it is determined at 1460 whether the histogram is normal. Depending on various configurations, a normal histogram occurs when the histogram peak and average value are close to each other within a threshold. If it is determined at 1460 that the histogram is not normally distributed, then the brightness controls of all illumination sources for each spectrum are adjusted by the same amount so that the peak of the intensity histogram of each image is substantially aligned with the target intensity 1462, and the process is complete at 1470. If it is determined at 1460 that the histogram is normally distributed, then the brightness controls of all illumination sources for each spectrum are adjusted by the same amount so that the average point of the image of each spectrum is substantially aligned with the target intensity 1464, and the process is complete at 1470. Figure 15A and Figure 15B Examples of misaligned and aligned normal histograms are shown respectively. Figure 16A and Figure 16B Examples of misaligned nonnormal histograms and aligned nonnormal histograms according to the embodiments described herein are shown respectively.

[0068] According to the various embodiments described herein, a flat-field normalization process can be used to further enhance image quality for colony counting. Using a flat-field normalization process allows for a substantially consistent response to the light level on the growth medium at the pixel scale used for colony counting (colony calculation). Reflection, diffusion, variability, etc., of the camera response are fixed-pattern variations that affect the image quality of colony counting. Flat-field normalization compensates for all fixed-pattern variations by mathematically smoothing the image to remove those caused by the imaging system.

[0069] The flat-field normalization process can be initiated by creating an average of multiple images illuminated with a previously calibrated light source brightness according to (1). Depending on the implementation, one or more high and / or low values ​​are discarded, and then the remaining values ​​of that pixel from the multiple images are averaged.

[0070]

[0071] Here, m and n are offsets in a two-dimensional array representing intensity values ​​in a 2D image. m is the horizontal offset, and n is the vertical offset, typically from the top left corner. avg [m,m] is the average pixel value at points m and n from multiple images. L is the average number of images, and I l These are the intensity values ​​at pixels m and n of image l in this set of images. Once the average value is calculated, the flat field, or in other words, the smoothed image value, will be calculated using the quadratic equation shown in (2).

[0072] I q [m, n]=β1+β2·m+β3·n+β4·m·n+β5·m 2 +β6·n 2 (2)

[0073] Here, I q These are the smoothed image values ​​at pixel locations m and n as defined above. These smoothed values ​​are the result of a uniform or flat response. The β values ​​are the coefficients of two bivariate second-order equations describing the smoothed image values ​​at pixel locations m and n. Depending on the configuration, (2) is a second-order curve fit used to make each pixel have the flattest response (e.g., the same response) to the set brightness level setting.

[0074] The β value in (2) can be determined by minimizing the squared error in the fit according to the equation in (3), where I is I avg And I q The flat field value is calculated from the above equation.

[0075]

[0076] Here, sq_err is the squared error between the average value at each pixel and the smoothed value calculated using (2). Depending on various configurations, sq_err is minimized to select the aforementioned β value. M is the number of pixel dimensions horizontally traversing the image, and N is the number of pixel dimensions vertically downwards along the image. Variables m and n are offsets defining the pixel positions. avg [m,n] is the average value at pixels m and n, and I q [m,n] are values ​​calculated using the measured β values ​​from the previous equation. According to various implementations, sq-err calculates the squared difference between the mean value and the flat field value for each pixel in the image. Minimizing this squared error yields a second-order equation describing the flat field value.

[0077] Once the value of the quadratic equation β is defined, the flat field gain used to multiply each pixel to achieve a flat field (smoothing value) can be calculated, as shown in (4).

[0078]

[0079] Here, variables m and n are the offsets of pixels in the aforementioned two-dimensional image. q [m,n] is an equation with β values ​​calculated at positions m and n in the image. I[m,n] is the average pixel intensity at positions m and n in the image. GainFlatField[m,n] is the ideal I... q The ratio to the average value at pixel [m,n], and indicates how much any pixel intensity value should be multiplied to adjust for a flat field smoothing response.

[0080] To further fine-tune the gain to adjust the flat field gain value and thus achieve device-to-device consistency, (5) can be applied to determine the total gain. This allows for scaling the smooth image gain so that the resulting adjusted response will achieve the desired target intensity.

[0081]

[0082] The process of generating a flat-field normalization factor may involve using an optical card representing the desired color balance and brightness. It may be desirable to use an optical card validated by industry-standard instruments. A normalization factor is generated for each color channel (spectrum). For example, a normalization factor can be generated for each of the red, blue, and green channels in a system utilizing these three spectra.

[0083] Figure 17 The process shown can be used to generate a normalization factor. Although Figure 17The example shown describes a system with red, blue, and green channels; however, it should be understood that this procedure can be applied to any number of color channels and / or color channels different from those listed herein. The instrument for the lighting system has already been calibrated using a 1710. The lighting system can be calibrated according to the implementation discussed herein and / or using one or more different procedures. All color channels are correctly set using the calibration procedure. In this example, all three channels (red, blue, and green) are set.

[0084] Capture monochrome images of the optical card for each of the 1720 color channels. Combine all monochrome images from the color channels. This step can be skipped if only one color channel is used. In this example, the monochrome images are combined into an RGB24 image. The steps of capturing the 1720 monochrome images and combining the 1730 monochrome images are repeated multiple times. In this example, multiple RGB24 images of a white card are created. The white card can be a medium (e.g., paper) with substantially balanced or near-balanced reflectance for each spectrum being calibrated. If the medium is substantially unbalanced, the amount of unbalance can be compensated to align each spectrum. As described herein, the region of interest (ROI) is shown at 1750 as the area for bacterial growth on the culture medium. While this example uses bacterial growth as the ROI, it should be understood that other types of growth, either in conjunction with or as an alternative to bacterial growth, can be used to define the ROI.

[0085] Use (6) to calculate the normalization factor for each of the 1760 images.

[0086] For the exponent within the range (0, number of images):

[0087]

[0088] As shown in (7), calculate the average value of the 1770 normalization factor.

[0089] Norm_factor=mean(norm_factor,axis=0) (7)

[0090] The average value calculated from (7) can be used as an instrument normalization factor, and it will be used to normalize all images captured by the instrument.

[0091] According to the various implementations described herein, normalization of each captured image can be performed by dividing each pixel value of the captured image by the corresponding pixel of the normalization factor. Figure 18An example is shown of how normalization can be used to generate a uniform image on a 5×5 sliding window based on each pixel. In this example, an intensity value 1810 corresponding to each pixel in the captured image is shown. Each pixel is divided by its corresponding normalization factor 1820 to create an output image 1830 with normalized intensity values.

[0092] Figure 19A and Figure 19B Exemplary images are shown using only the calibration illumination system (19A) described herein and using the calibration illumination system in addition to the flat field normalization process. It can be observed that, compared to... Figure 19A Compared to the uneven and non-uniform ROI demonstrated by the darker areas on the left and right edges, Figure 19B The central region of the ROI in the image is basically uniform across the entire range.

[0093] Figure 20 An exemplary histogram according to the embodiments described herein is shown, illustrating the light distribution of each red, green, and blue channel and the effect of flat-field normalization on image uniformity. Before flat-field normalization, the pixel intensity distribution ranges from approximately 102 to approximately 230. After flat-field normalization, the pixel intensity distribution decreases to the range of approximately 155 to approximately 225. According to various embodiments, the pixel intensity distribution decreases to the range of approximately 178 to approximately 220.

[0094] Figures 21A to 21F Heatmap representations of each of the red, green, and blue channels before and after normalization according to the implementation described herein are shown. Specifically, Figure 21A The red channel heatmap before normalization is shown. Figure 21B The normalized red channel heatmap is shown. Figure 21C The blue channel heatmap before normalization is shown. Figure 21D The normalized blue channel heatmap is shown. As can be observed from this example, in Figure 21D After normalization, the non-uniformity at the corners is essentially eliminated in the heatmap 2150. Figure 21E The green channel heatmap before normalization is shown. Figure 21F The normalized green channel heatmap is shown. As can be observed from this example, in Figure 21D After normalization, the non-uniformity at the corners is essentially eliminated in the heatmap 2170.

[0095] The above methods can be implemented on a computer using well-known computer processors, memory units, storage devices, computer software, and other components. Figure 22A high-level block diagram of such a computer is shown. Computer 2200 includes a processor 2210 that controls the overall operation of computer 2200 by executing computer program instructions that define such operations. It should be understood that processor 2210 may include any type of device capable of executing instructions. For example, processor 2210 may include one or more of a central processing unit (CPU), a graphics processing unit (GPU), a field-programmable gate array (FPGA), and an application-specific integrated circuit (ASIC). The computer program instructions may be stored in storage device 2220 (e.g., a disk) and loaded into memory 2230 when the computer program instructions are desired to be executed. Therefore, the steps of the methods described herein may be defined by computer program instructions stored in memory 2230 and controlled by processor 2210 that executes the computer program instructions. Computer 1500 may include one or more network interfaces 2250 for communicating with other devices via a network. Computer 2200 also includes a user interface 2260 that enables a user to interact with computer 1500. User interface 2260 may include I / O devices 2262 (e.g., keyboard, mouse, speaker, buttons, etc.) to allow a user to interact with the computer. Such input / output devices 2262 may be used in conjunction with a set of computer programs according to embodiments described herein. The user interface also includes a display 2264. According to various embodiments, Figure 22 It is a high-level representation of possible components of a computer for illustrative purposes, and a computer may contain other components.

[0096] Exemplary Implementation

[0097] Implementation Plan 1. A method comprising:

[0098] Identify the region of interest on the growth medium;

[0099] Calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest;

[0100] Based on the total brightness control value, an individual brightness value for each of the plurality of lighting sources is calculated by individually adjusting the brightness of each lighting source to generate at least one image, the at least one image substantially matching the target intensity in the influence area of ​​each corresponding lighting source; and

[0101] The calibrated brightness value for each illumination source is determined based on the following: the image intensity when each illumination source is turned on with its respective individual brightness value, and the intensity generated by each illumination source in each corresponding influence area when turned on individually.

[0102] Implementation Scheme 2. The method according to any one of Implementation Schemes 1 and 3-11 further includes performing the steps according to Implementation Scheme 1 for each of the plurality of spectra.

[0103] Implementation Scheme 3. The method according to any one of Implementation Schemes 1, 2 and 4-11 further includes:

[0104] Multiple images of the region of interest are captured using the calibrated brightness values;

[0105] A pixel histogram of the intensity of each image is generated based on the plurality of images, with each of the plurality of brightness values.

[0106] Determine the average and peak values ​​of the pixel histogram;

[0107] The calibrated brightness value is adjusted based on at least one of the average value and the peak value of the pixel histogram.

[0108] Implementation Scheme 4. The method according to any one of Implementation Schemes 1-3 and 5-11 further includes using one or more of linear regression, second-order curve fitting, multi-point averaging, and curve smoothing to calculate at least one of the total brightness control value and the individual brightness value.

[0109] Implementation Scheme 5. The method according to any one of Implementation Schemes 1-4 and 6-11, wherein calculating the total brightness control value includes:

[0110] Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity;

[0111] Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity.

[0112] Calculate the average intensity of each captured image; and

[0113] The total brightness control value is determined based on the average value.

[0114] Implementation Scheme 6. The method according to any one of Implementation Schemes 1-5 and 7-11, wherein calculating the individual luminance value comprises:

[0115] For each of the plurality of lighting sources:

[0116] Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the total brightness control value.

[0117] Calculate the average intensity of each captured image; and

[0118] The individual brightness value is determined based on the average value.

[0119] Implementation Scheme 7. The method according to any one of Implementation Schemes 1-6 and 8-11, wherein the area of ​​interest includes a bacterial growth zone.

[0120] Implementation Scheme 8. The method according to any one of Implementation Schemes 1-7 and 9-11 further includes:

[0121] The normalization factor is calculated using the flat-field normalization process; and

[0122] The calibration brightness value is adjusted based on the normalization factor.

[0123] Implementation Scheme 9. The method according to Implementation Scheme 8, wherein the flat field normalization process includes:

[0124] A monochrome image is captured using the calibrated brightness value for each of the multiple spectra; and

[0125] The normalization factor for each spectrum in the spectrum is calculated using the monochrome image.

[0126] Implementation Scheme 10. The method according to any one of Implementation Schemes 1-9 and 11, wherein determining the calibrated brightness value comprises:

[0127] Calculate the difference between the target intensity and the image intensity at each individual brightness value;

[0128] Calculate the percentage change for each individual brightness value to compensate for the difference;

[0129] Capture multiple images of the region of interest;

[0130] Calculate the average intensity of each captured image; and

[0131] The calibrated brightness value is determined based on the average value and the percentage change value.

[0132] Implementation Scheme 11. The method according to any one of Implementation Schemes 1-10 further includes:

[0133] For each of the plurality of lighting sources:

[0134] Capture at least one image while the illumination source is on; and

[0135] The affected region is determined by identifying the image region where the image intensity is a threshold percentage of the target intensity.

[0136] Implementation Scheme 12. A system comprising:

[0137] processor; and

[0138] A memory that stores computer program instructions, which, when executed by the processor, cause the processor to perform operations including:

[0139] Identify the region of interest on the growth medium;

[0140] Calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest;

[0141] Based on the total brightness control value, an individual brightness value for each of the plurality of lighting sources is calculated by individually adjusting the brightness of each lighting source to generate at least one image, the at least one image substantially matching the target intensity in the influence area of ​​each corresponding lighting source; and

[0142] The calibrated brightness value for each illumination source is determined based on the following: the image intensity when each illumination source is turned on with its respective individual brightness value, and the intensity generated by each illumination source in each corresponding influence area when turned on individually.

[0143] Implementation Scheme 13. The system according to any one of Implementation Schemes 12 and 14-19, wherein the lighting source comprises one or more of a light-emitting diode (LED), an incandescent device, and a fluorescent device.

[0144] Implementation Scheme 14. The system according to any one of Implementation Schemes 12, 13 and 15-19, wherein the processor is further configured to:

[0145] Multiple images of the region of interest are captured using the calibrated brightness values;

[0146] Based on the plurality of images, a pixel histogram of the intensity of each image is generated for each of the plurality of brightness values; and

[0147] Determine the average and peak values ​​of the pixel histogram.

[0148] Implementation Scheme 15. The system according to any one of Implementation Schemes 12-14 and 16-19, wherein the processor is further configured to use linear regression to calculate at least one of the total brightness control value and the individual brightness value.

[0149] Implementation Scheme 16. The system according to any one of Implementation Schemes 12-15 and 17-19, wherein calculating the total brightness control value includes:

[0150] Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity;

[0151] Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity.

[0152] Calculate the average intensity of each captured image; and

[0153] The total brightness control value is determined based on the average value.

[0154] Implementation Scheme 17. The system according to any one of Implementation Schemes 12-16, 18 and 19, wherein calculating the individual luminance value comprises:

[0155] For each of the plurality of lighting sources:

[0156] Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the brightness control value.

[0157] Calculate the average intensity of each captured image; and

[0158] The individual brightness value is determined based on the average value.

[0159] Implementation Scheme 18. The system according to any one of Implementation Schemes 12-17 and 19, wherein the area of ​​interest includes a bacterial growth zone.

[0160] Implementation Scheme 19. The system according to any one of Implementation Schemes 12-18, wherein the processor is further configured to:

[0161] The normalization factor is calculated using the flat-field normalization process; and

[0162] The calibration brightness value is adjusted based on the normalization factor.

[0163] Implementation Scheme 20. A non-transitory computer-readable medium storing computer program instructions for designing microstructures, the computer program instructions, when executed by a processor, causing the processor to perform operations including:

[0164] Identify the region of interest on the growth medium;

[0165] Calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest;

[0166] Based on the total brightness control value, an individual brightness value for each of the plurality of lighting sources is calculated by individually adjusting the brightness of each lighting source to generate at least one image, the at least one image substantially matching the target intensity in the influence area of ​​each corresponding lighting source; and

[0167] The calibrated brightness value for each illumination source is determined based on the following: the image intensity when each illumination source is turned on with its respective individual brightness value, and the intensity generated by each illumination source in each corresponding influence area when turned on individually.

[0168] Unless otherwise specified, all figures used in the specification and claims to express structural dimensions, quantities, and physical properties should in all cases be understood to be modified by the term "about". Therefore, unless stated to the contrary, the numerical parameters listed in the foregoing specification and appended claims are approximations, which may vary according to the desired properties sought by a person skilled in the art using the teachings disclosed herein. The use of numerical ranges expressed by endpoints includes all numbers within that range (e.g., 1 to 5 includes 1, 1.5, 2, 2.75, 3, 3.80, 4, and 5) and any range within that range.

[0169] The various implementation schemes described above can be implemented using circuitry and / or software modules that provide specific results in an interactive manner. Those skilled in the art can readily implement the functions described in this way, either at the module level or as a whole, using knowledge well-known in the art. For example, the flowcharts shown herein can be used to create computer-readable instructions / code for processor execution. Such instructions can be stored on a computer-readable medium and transmitted to the processor for execution in accordance with methods well-known in the art. The structures and procedures shown above are merely representative examples of implementation schemes that can be used to facilitate the above-described implementation schemes.

[0170] The above description of exemplary embodiments is for illustrative and descriptive purposes only. It is not intended to exhaustively describe the concept of the invention or to limit the concept of the invention to the explicit forms disclosed herein. Many modifications and variations can be used in accordance with the above teachings. Any or all features of the disclosed embodiments may be applied individually or in any combination, and are not intended to be limiting but are purely illustrative. The scope of the invention is defined by the appended claims, and not by the detailed description.

Claims

1. A method comprising: Use imaging devices to identify regions of interest on the growth medium; Calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest; Based on the total brightness control value, an individual brightness value for each of the plurality of lighting sources is calculated by individually adjusting the brightness of each lighting source to generate at least one image, the at least one image substantially matching the target intensity in the influence area of ​​each corresponding lighting source; and The calibrated brightness value for each illumination source is determined based on the following: the image intensity when each illumination source is turned on with its respective individual brightness value, and the intensity generated by each illumination source in each corresponding area of ​​influence when turned on individually; Multiple images of the region of interest are captured at the calibrated brightness value, and the average pixel intensity of each image is captured. The calculation of the total brightness control value includes: Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity; Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity. Calculate the average intensity of each captured image; and The total brightness control value is determined based on the average value; The calculation of the individual brightness value includes: For each of the plurality of lighting sources: Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the total brightness control value. Calculate the average intensity of each captured image; and The individual brightness value is determined based on the average value; Determining the calibration brightness value includes: Calculate the difference between the target intensity value and the image intensity at each individual brightness value; Calculate the percentage change for each individual brightness value to compensate for the difference; Capture multiple images of the region of interest; Calculate the average intensity of each captured image; and The calibrated brightness value is determined based on the average value and the percentage change value.

2. The method of claim 1, further comprising performing the steps of claim 1 for each of the plurality of spectra.

3. The method according to claim 1, further comprising: A pixel histogram of the intensity of each image is generated based on the plurality of images, with each of the plurality of brightness values. Determine the average and peak values ​​of the pixel histogram; as well as The calibrated brightness value is adjusted based on at least one of the average value and the peak value of the pixel histogram.

4. The method of claim 1, further comprising using one or more of linear regression, second-order curve fitting, multi-point averaging, and curve smoothing to calculate at least one of the total brightness control value and the individual brightness value.

5. The method according to claim 1, wherein the region of interest includes a bacterial growth zone.

6. The method according to claim 1, further comprising: The normalization factor is calculated using the flat-field normalization process; as well as The calibration brightness value is adjusted based on the normalization factor.

7. The method according to claim 6, wherein the flat-field normalization process comprises: A monochrome image is captured using the calibrated brightness value for each of the multiple spectra; as well as The normalization factor for each spectrum in the spectrum is calculated using the monochrome image.

8. The method according to claim 1, further comprising: For each of the plurality of lighting sources: Capture at least one image while the lighting source is on; as well as The affected region is determined by identifying the image region where the image intensity is a threshold percentage of the target intensity.

9. A system comprising: processor; and A memory that stores computer program instructions, which, when executed by the processor, cause the processor to perform operations including: The processor is used to determine the region of interest on the growth medium; The processor is configured to calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest; The processor is configured to calculate an individual brightness value for each of the plurality of lighting sources based on the total brightness control value by individually adjusting the brightness of each lighting source to generate at least one image that substantially matches the target intensity in the influence area of ​​each corresponding lighting source. and The processor is configured to determine a calibrated brightness value for each illumination source based on: the image intensity of each illumination source when turned on with a corresponding individual brightness value, and the intensity generated by each illumination source in each corresponding area of ​​influence when turned on individually; The processor is configured to capture multiple images of the region of interest at the calibrated brightness value and capture the average pixel intensity of each image; The calculation of the total brightness control value includes: Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity; Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity. Calculate the average intensity of each captured image; and The total brightness control value is determined based on the average value; The calculation of the individual brightness value includes: For each of the plurality of lighting sources: Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the total brightness control value. Calculate the average intensity of each captured image; and The individual brightness value is determined based on the average value; Determining the calibration brightness value includes: Calculate the difference between the target intensity value and the image intensity at each individual brightness value; Calculate the percentage change for each individual brightness value to compensate for the difference; Capture multiple images of the region of interest; Calculate the average intensity of each captured image; and The calibrated brightness value is determined based on the average value and the percentage change value.

10. The system of claim 9, wherein the lighting source comprises one or more of a light-emitting diode (LED), an incandescent device, and a fluorescent device.

11. The system of claim 9, wherein the processor is further configured to: Multiple images of the region of interest are captured using the calibrated brightness values; Based on the plurality of images, a pixel histogram of the intensity of each image is generated for each of the plurality of brightness values; and Determine the average and peak values ​​of the pixel histogram.

12. The system of claim 9, wherein the processor is further configured to use linear regression to calculate at least one of the total brightness control value and the individual brightness value.

13. The system of claim 9, wherein calculating the total brightness control value comprises: Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity; Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity. Calculate the average intensity of each captured image; and The total brightness control value is determined based on the average value.

14. The system of claim 9, wherein calculating the individual brightness value comprises: For each of the plurality of lighting sources: Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the total brightness control value. Calculate the average intensity of each captured image; and The individual brightness value is determined based on the average value.

15. The system of claim 9, wherein the region of interest includes a bacterial growth zone.

16. The system of claim 9, wherein the processor is further configured to: The normalization factor is calculated using the flat-field normalization process; and The calibration brightness value is adjusted based on the normalization factor.

17. A non-transitory computer-readable medium storing computer program instructions for designing microstructures, the computer program instructions, when executed by a processor, causing the processor to perform operations including: The processor is used to determine the region of interest on the growth medium; The processor is configured to calculate a total brightness control value for a plurality of illumination sources configured to irradiate the growth medium, the total brightness control value generating at least one image that substantially matches the target intensity at the region of interest; The processor is configured to calculate an individual brightness value for each of the plurality of lighting sources based on the total brightness control value by individually adjusting the brightness of each lighting source to generate at least one image that substantially matches the target intensity in the influence area of ​​each corresponding lighting source. and The processor is configured to determine a calibrated brightness value for each illumination source based on: the image intensity of each illumination source when turned on with a corresponding individual brightness value, and the intensity generated by each illumination source in each corresponding area of ​​influence when turned on individually; The processor is configured to capture multiple images of the region of interest at the calibrated brightness value and capture the average pixel intensity of each image; The calculation of the total brightness control value includes: Simultaneously adjust the brightness of all the multiple lighting sources to a level lower than the target intensity; Multiple images of the region of interest are captured iteratively, starting with a brightness level that increases from below the target intensity. Calculate the average intensity of each captured image; and The total brightness control value is determined based on the average value; The calculation of the individual brightness value includes: For each of the plurality of lighting sources: Multiple images of the region of interest are captured iteratively at brightness levels that increase from below the total brightness control value. Calculate the average intensity of each captured image; and The individual brightness value is determined based on the average value; Determining the calibration brightness value includes: Calculate the difference between the target intensity value and the image intensity at each individual brightness value; Calculate the percentage change for each individual brightness value to compensate for the difference; Capture multiple images of the region of interest; Calculate the average intensity of each captured image; and The calibrated brightness value is determined based on the average value and the percentage change value.

Images