Inspection Equipment

Abstract

To provide a test device capable of determining reliability of an image feature quantity in a target region of an observation image of a sample.SOLUTION: A test device of the present invention is configured to extract a first feature quantity from a target region, including a test target, of an observation image of a sample, extract a second feature quantity from a reference region, other than the target region, of the observation image, and compare the first feature quantity and the second feature quantity to compute reliability of the first feature quantity.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to an inspection apparatus that inspects a sample using an observation image of the sample.

Background Art

[0002] A method has been reported in which an image of a sample observed using an apparatus such as a microscope is acquired and an inspection is performed based on the image. Non-Patent Document 1 below describes a method for inspecting the number of particles at an inspection position by comparing the image feature amount at the inspection position with the feature amount (number of particles) of the background portion among the observation images acquired by an electron microscope.

Prior Art Documents

Non-Patent Documents

[0003]

Non-Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When using the above inspection method, if there is no significant difference between the image feature amount at the inspection position and the image feature amount at the reference position, it becomes difficult to accurately specify the number of particles at the inspection position. As a result, the inspection accuracy decreases.

[0005] In order to show that the difference is significant, it is also conceivable to acquire a plurality of observation images and compare the feature amounts of the inspection position and the reference position in each observation image. However, even if a plurality of observation images are acquired, they are not necessarily all suitable for comparison. For example, due to factors such as the sample preparation process, impurities may be mixed into the inspection position or the reference position, resulting in the generation of defective images. Using such defective images will reduce the reliability of the image feature amount, and a highly reliable result cannot be obtained in the inspection using the feature amount of the image. If the number of images is significantly increased, the reliability of the inspection result will be improved to a certain extent, but on the other hand, the inspection throughput will decrease.

[0006] The present invention has been made in view of the above problems, and an object thereof is to provide an inspection apparatus capable of determining the reliability of an image feature amount in a target region of an observation image of a sample.

Means for Solving the Problems

[0007] The inspection apparatus according to the present invention extracts a first feature amount from a target region including an inspection target in an observation image of a sample, extracts a second feature amount from a reference region other than the target region in the observation image, and calculates the reliability of the first feature amount by comparing the first feature amount and the second feature amount.

Effects of the Invention

[0008] According to the inspection apparatus according to the present invention, the reliability of the image feature amount in the target region of the observation image of the sample can be determined. Other problems, configurations, advantages, etc. of the present invention will become clear by referring to the following embodiments.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Mode for Carrying Out the Invention

[0010] <Embodiment 1> FIG. 1 is a configuration diagram of an inspection apparatus 1 according to Embodiment 1 of the present invention. The inspection apparatus 1 is an apparatus that inspects a sample using an observation image of the sample. The inspection apparatus 1 includes an image acquisition device 10, an evaluation device 20, and an input / output device 30. As a sample, for example, an immunochromatography test kit is used. The immunochromatography test kit is a device that detects an antigen or an antibody by flowing a sample solution containing the antigen or antibody to be inspected onto a plate member formed of a porous body or the like. The marker particles contained in the sample solution or the plate-like member flow on the surface or inside of the plate-like member together with the sample solution, and the antigen or antibody can be visually detected by the color change caused by the marker particles captured by the capture antibody in the plate-like member. Examples of the marker particles include various conductive particles and insulating particles such as metal particles, latex particles, and silica particles. It is also possible to perform an inspection using the feature amount of the image of the test kit instead of the color change caused by the marker particles. For example, the feature amount at the inspection position is compared with the feature amount at another reference position. The reference position is a position where it is assumed that no (or very few) marker particles are present. By comparing the feature amount at the inspection position with the feature amount at the reference position as a reference, the number of marker particles (that is, the amount of antigen or antibody) at the inspection position can be inspected. As a sample, for example, in an immunochromatography test kit, a plate-like member that holds a liquid containing marker particles can be considered, but it is not limited thereto.

[0011] The image acquisition device 10 acquires an observation image of the sample. Instead of inspecting the sample by visually checking it, the sample is inspected by analyzing the image feature amount of the sample. For example, by evaluating the image feature amount correlated with the number of marker particles, a more accurate inspection result can be obtained than visually checking the marker particles. The specific configuration of the image acquisition device 10 will be described later.

[0012] The evaluation device 20 evaluates the reliability of the feature amount of the observation image acquired by the image acquisition device 10. The reliability here refers to an index indicating whether there is a significant difference between the image feature amount of the target area where the inspection target (marker particles in the case of an immunochromatography test kit) exists and the image feature amount of the reference area (an area on the test kit where it is assumed that no marker particles exist or the amount is very small). Examples of significant differences and reliability will be described later.

[0013] The evaluation device 20 includes an image correction unit 21, a feature amount extraction unit 22, a target area specification unit 23, an occupancy rate analysis unit 24, a feature amount comparator 25, a reliability evaluation unit 26, a re-acquisition determination unit 27, and a storage unit 28. The storage unit 28 can be configured by a storage device that stores data. Other functional units can also be configured by hardware such as circuit devices that implement these functions, or can be configured by software that implements these functions being executed by an arithmetic device such as a CPU (Central Processing Unit). The operations of these functional units will be described later.

[0014] The input / output device 30 is a device that displays the processing result by the evaluation device 20 or inputs an instruction given by the user to the evaluation device 20 and the image acquisition device 10. The input / output device 30 includes an image display unit 31, a selection unit 32, a reliability display unit 33, and a feature amount display unit 34. The operations of these functional units will be described later.

[0015] FIG. 2 is a diagram for explaining a significant difference in an immunochromatography test kit. In the immunochromatography test kit (sample) 40, an inspection position 42 (target area) and a reference position 43 (reference area) are set on a porous plate-like member 41. A liquid containing an inspection target such as an antigen or an antibody is flowed in the direction of the arrow on the plate-like member 41. When the inspection target binds to marker particles such as metal particles, the number of the inspection targets can be inspected by counting the marker particles.

[0016] In this embodiment, in order to improve the inspection accuracy, instead of counting the marker particles themselves, the amount of the inspection target at the inspection position 42 is determined by comparing the image feature amount at the inspection position 42 with the image feature amount at the reference position 43. The reference position 43 is assumed to have no inspection target particles or a very small amount. The image feature amount is the number of inspection target particles or an amount corresponding thereto. For example, the area, density, etc. of the inspection target particles can be mentioned.

[0017] The actual image feature amount at the reference position 43 is not necessarily 0 and may vary for each observed image. As a result, there may be no significant difference in the image feature amount between the inspection position 42 and the reference position 43. In the example shown in FIG. 2, the first observed image at the inspection position 42 contains 20 inspection target particles, and the first observed image at the reference position 43 contains 10 inspection target particles. The second and subsequent observed images also contain inspection target particles as shown in FIG. 2. If it cannot be said that these feature amounts have a significant difference, when inspecting the observed image at the inspection position 42 with the observed image at the reference position 43 as a reference, there is a possibility that a correct inspection result cannot be obtained.

[0018] Therefore, in the first embodiment, the reliability of the image feature amount at the inspection position 42 is determined by evaluating whether there is a significant difference between the image feature amount at the inspection position 42 and the image feature amount at the reference position 43.

[0019] FIG. 3 is a side sectional view showing a configuration example of the image acquisition device 10. The image acquisition device 10 is a device that acquires an observed image of the sample 40 by irradiating the sample 40 with, for example, an electron beam, and is a type of charged particle beam device. The image acquisition device 10 includes an image acquisition control unit 11, an image formation unit 12, a stage control unit 13, an electron source 14, a deflector 15, a lens 16, a stage 17, and a detector 18.

[0020] The electron source 14 emits an electron beam. The deflector 15 deflects the direction of the electron beam. The lens 16 irradiates the electron beam onto the sample 40. The sample 40 is placed on the stage 17. The detector 18 detects secondary particles generated from the sample 40 by irradiating the electron beam onto the sample 40, and outputs a detection signal representing the intensity thereof. The image forming unit 12 generates an observation image of the sample 40 using the detection signal. The stage control unit 13 controls the stage 17. The image acquisition control unit 11 controls the overall operation of the image acquisition device 10.

[0021] FIG. 4 is a flowchart for explaining a conventional procedure for performing an inspection using an observation image of an immunochromatography test kit. In the conventional procedure, observation images are acquired at each of the reference position 43 and the inspection position 42. When a prescribed number of observation images are obtained, the feature amounts at each position are compared with each other and the result is analyzed. This series of procedures is performed by a person operating the apparatus. In this procedure, it is not evaluated whether there is a significant difference between the image feature amount at the reference position 43 and the image feature amount at the inspection position 42. Therefore, there is a possibility that the inspection result may not be reliable.

[0022] FIG. 5 is a flowchart for explaining the operation of the inspection apparatus 1. This flowchart includes an operation in which the inspection apparatus 1 determines the reliability of the image feature amount at the inspection position 42. Each step of FIG. 5 will be described below.

[0023] (FIG. 5: Step S501) The stage control unit 13 moves the stage 17 to a position where the electron beam is irradiated onto the reference position 43. The image acquisition control unit 11 adjusts imaging conditions such as optical conditions. The image acquisition control unit 11 controls each unit so as to irradiate the sample 40 with the electron beam. By controlling each unit, adjustments such as the setting of the detector for forming an image, focusing of the electron beam, and scanning method are performed. The image forming unit 12 generates an observation image at the reference position 43.

[0024] (FIG. 5: Step S502) The target area specifying unit 23 specifies an area (referred to as the target area) in the observation image acquired in S501 for which an inspection is to be performed using feature amounts. In the observation image, there are portions that are unnecessary when calculating the feature amounts of the inspection target. In this step, the image area obtained by removing such unnecessary portions from the observation image is specified as the target area. A specific example of this step will be described later.

[0025] (Fig. 5: Step S503) The image correction unit 21 corrects the image quality of the observation image. For example, the contrast of the entire observation image is enhanced so that the feature amounts of the inspection target particles are more emphasized. This step may be performed before S502 (a configuration example of this is shown in Fig. 1). Specific examples of this step will be described later.

[0026] (Fig. 5: Step S504) The feature amount extraction unit 22 extracts the image feature amounts of the target area specified in S502. The image feature amount may be the number of inspection target particles, or a numerical value having an equivalent meaning. Specific examples of the feature amounts will be described later. For example, the image feature amounts can be calculated by extracting the inspection target particles from the observation image by image segmentation. Additionally, for example, the image feature amounts may be calculated from statistical amounts such as the histogram of the luminance values of the pixels. Other appropriate methods may also be used.

[0027] (Fig. 5: Steps S505 - S508) The inspection device 1 extracts the image feature amounts of the target area in the same manner as S501 - S504 for the inspection position 42.

[0028] (Fig. 5: Step S509) The feature amount comparator 25 determines whether there is a significant difference between the image feature amounts at the reference position 43 and the image feature amounts at the inspection position 42 by comparing them. As an evaluation value that can be used to determine whether there is a significant difference, for example, the P - value of a t - test can be considered. Calculate the P - value between the two feature amounts, and if the P - value is sufficiently small (for example, P - value < 0.01), it can be considered that there is a significant difference.

[0029] (Figure 5: Step S509: Calculation Example) In the example shown in FIG. 2, using the P-value between the set of six feature amounts acquired at the reference position 43 and the set of six feature amounts acquired at the inspection position 42, it is determined whether or not there is a significant difference.

[0030] (Figure 5: Step S510) The reliability evaluation unit 26 calculates the reliability of the image feature amounts at the inspection position 42 according to the result of S509. A calculation procedure is used such that the greater the significant difference, the higher the reliability according to the parameters calculated in S509. However, it is not always necessary to use a continuous quantity as the reliability. For example, if the significant difference is equal to or greater than the threshold value, the reliability may be set to a high fixed value (for example, 1), and if it is less than the threshold value, it may be set to a low fixed value (for example, 0). That is, any form of reliability may be used as long as it can be determined whether or not the image feature amounts at the inspection position 42 are suitable for the inspection.

[0031] (Figure 5: Step S511) The re-acquisition determination unit 27 determines whether or not it is necessary to re-acquire the observation image by comparing the reliability calculated in S510 with the threshold value. If the reliability is equal to or greater than the threshold value, the process proceeds to S513, and if it is less than the threshold value, the process proceeds to S512. This step has the significance of providing a limit so as to acquire only the necessary amount of observation images by stopping the acquisition of the observation image when the reliability reaches the threshold value or more.

[0032] (Figure 5: Step S512) The re-acquisition determination unit 27 determines whether or not it is necessary to re-acquire the observation image by checking the number of acquired observation images. If the number of acquired observation images is equal to or greater than the predetermined number, the process proceeds to S513, and otherwise, the process returns to S501 to re-acquire the observation image. This step has the significance of restricting so as not to acquire an excessive amount of observation images even if sufficient reliability cannot be obtained.

[0033] (Figure 5: Step S513) The reliability display unit 33 displays the reliability calculated in S510. An example of the screen display will be described later.

[0034] (Fig. 5: S501~S508: Supplementary Note 1) In this flowchart, the observation image at the reference position 43 is acquired first, and then the observation image at the inspection position 42 is acquired. However, the observation image at the inspection position 42 may be acquired first, and then the observation image at the reference position 43 may be acquired. In this case, S505~S508 are performed first, and then S501~S504 are performed. Only the observation images at each position may be acquired first, and the feature quantity extraction may be performed after each observation image is acquired. The same applies to Embodiment 2.

[0035] (Fig. 5: S501~S508: Supplementary Note 2) In the immunochromatography test kit, it is desirable that the state of the inspection position 42 is as stable as possible. Therefore, by first acquiring the observation image of the reference position 43, sufficient drying time may be ensured so that the state of the inspection position 42 becomes stable. The same applies to Embodiment 2. In this case, S501~S508 are performed in the order as described in Fig. 5.

[0036] Fig. 6 shows a specific example of S503. When the image contrast between the target region 121 and the inspection target particle 122 is low, the image correction unit 21 adjusts the contrast of the entire observation image. As a result, as shown in the right figure of Fig. 6, the contrast between the target region 121 and the inspection target particle 122 becomes clear, and the feature quantity of the inspection target particle 122 can be acquired more accurately. It is desirable to correct both the inspection position 42 and the reference position 43. However, for example, a case where the contrast of the other is corrected based on one of them is also conceivable. Also, when correcting the contrast of the observation image, correction considering the irradiation conditions of the electron beam and the influence of the irradiation conditions on the image may be performed.

[0037] FIG. 7 is a schematic diagram for explaining a specific example of S502. In the observation image, in addition to the inspection target particle 122, there may be portions unnecessary for calculating the feature amount of the inspection target particle 122, such as a region 123 where the feature amount is hidden and a region 124 where there is no information. The target region specifying unit 23 specifies, as the target region 121, an image region obtained by removing these unnecessary portions from the observation image.

[0038] Region 123 is a closed region with a large luminance value and a large size. Examples of the closed region include, for example, large foreign matters. Region 124 is, for example, a void of the porous material included in the plate-like member 41 or an image artifact. Examples of the image artifact include charging due to electron beam irradiation, damage, morphological changes due to vacuum evacuation, and those due to mistakes in image luminance adjustment.

[0039] Region 123 is formed by a large foreign matter, and its luminance value may be relatively large or low compared to other regions in the observation image. In this embodiment, the case where the luminance value is large is shown. Region 124 has a smaller luminance value than other regions if it is a void, for example. Therefore, the target region 121 can be specified as those having a luminance value within a predetermined range (between the upper limit value and the lower limit value) in the observation image. The target region specifying unit 23 specifies the target region 121 by this method. The right figure of FIG. 7 shows the result of specifying the target region 121 by removing regions 123 and 124.

[0040] As another method for specifying the target region 121, it is conceivable to previously learn, by machine learning, unnecessary portions that are not the target region 121, such as regions 123 and 124, and remove the unnecessary portions using the learning result. For example, by learning feature amounts such as the size, shape, and luminance value of the unnecessary portions, these can be specified. The target region 121 may be specified by other appropriate methods.

[0041] As the image feature amount extracted by the feature amount extraction unit 22, a feature amount equivalent to the number may be extracted instead of or in combination with the number of inspection target particles 122. For example, feature amounts such as the area and density of the inspection target particles 122 can be considered. The ratio of these to the area of the target region 121 can also be used as a feature amount. The image feature amount in this case can be calculated by the following formula: Image feature amount = (Feature amount of inspection target particles 122) / (Area of target region 121). The feature amount of the inspection target particles 122 is, for example, the number and area of the inspection target particles 122. This feature amount is called the occupancy rate (derived feature amount) in the first embodiment. The occupancy rate may be calculated by the feature amount extraction unit 22, or a functional unit for calculating this may be provided like the occupancy rate analysis unit 24.

[0042] The feature amount extraction unit 22 extracts the image feature amount by the same method for the inspection position 42 and the reference position 43. For example, when using the above formula, the image feature amount is extracted by the above formula at each position, and these are compared in S509.

[0043] FIG. 8 is an example of a user interface provided by the input / output device 30. The feature amount display unit 34 displays the image feature amounts of the respective observation images acquired at the inspection position 42 and the image feature amounts of the respective observation images acquired at the reference position 43 in the display column 341. The reliability display unit 33 displays the reliability of the image feature amount of the observation image at the inspection position 42 in the display column 331. This user interface can be configured, for example, by the input / output device 30 acquiring each value calculated by the evaluation device 20 and displaying it on the screen.

[0044] FIG. 9 is an example of a user interface provided by the input / output device 30. The image display unit 31 displays each observation image acquired by the image acquisition device 10 as shown in FIG. 9. Information such as the target region 121 and its ratio may be presented together with the observation image.

[0045] FIG. 10 shows an example of data stored in the storage unit 28. The storage unit 28 stores each data acquired by the evaluation device 20. For example, an observation image of the sample 40, an image of the target region 121, a feature amount of the observation image, a reliability, and the like can be mentioned.

[0046] <Embodiment 1: Summary> The inspection device 1 according to the first embodiment calculates the reliability of the image feature amount of the inspection position 42 by comparing the image feature amount of the inspection position 42 with the image feature amount of the reference position 43. Thereby, it is possible to determine in advance whether or not the observation image of the inspection position 42 is suitable for inspecting the inspection target. Therefore, if it is not suitable for inspection, measures such as acquiring an observation image again can be taken.

[0047] The inspection device 1 according to the first embodiment calculates the reliability of the image feature amount of the inspection position 42 by calculating an evaluation value (for example, the P value of a t-test) indicating whether or not there is a significant difference between the image feature amount of the inspection position 42 and the image feature amount of the reference position 43. Thereby, an observation image that is not suitable as a reference image can be excluded. Therefore, the inspection accuracy can be improved.

[0048] <Embodiment 2> In the second embodiment of the present invention, another operation example of the inspection device 1 will be described. Since the configuration of the inspection device 1 is the same as that of the first embodiment, the difference in the operation procedure will be mainly described below, and the matters common to the first embodiment will be omitted.

[0049] FIG. 11 is a flowchart for explaining the operation of the inspection device 1 in the second embodiment. S1101 is added between S502 and S503, S1102 is added between S506 and S507, and S1103 is added before returning from S512 to S501. The other steps are the same as those in FIG. 5.

[0050] (FIG. 11: Steps S1101, S1102) The target area specifying unit 23 determines whether the size of the target area 121 specified in S502 is equal to or greater than a threshold value. If the size of the target area 121 is equal to or greater than the threshold value, the process proceeds to S503. If it is less than the threshold value, the process returns to S501 to re-acquire the observation image. Depending on the method of specifying the target area 121, the target area 121 may be too small. For example, this may occur when a very large foreign object is included in the observation image. If the target area 121 is too small, there is a possibility that the feature amount of the inspection target particle 122 cannot be appropriately acquired. Therefore, in such a case, the observation image is re-acquired. In S1102, the same process is performed for the inspection position 42.

[0051] (Fig. 11: Step S1103) The re-acquisition determination unit 27 determines whether the factor with low reliability is due to the observation image at the reference position 43. If the reference image is the factor, the process returns to S501 to re-acquire the observation image of the reference position 43. Otherwise, the process returns to S505 to re-acquire the observation image of the inspection position 42. When returning to S501, the inspection position 42 will also be re-acquired. Reasons why the reference image is the factor include, for example, the number of reference images not reaching the reference number, and the numerical value of the feature amount of the reference image being too large.

[0052] (Fig. 11: Step S1103: Supplementary) This flowchart is configured to first acquire a reference image and then acquire an inspection image. However, there may be a case where the reference image is acquired after the inspection image is acquired first. Even in this case, when it is determined in S1103 that the reference image is the factor, the process returns to the step of re-acquiring the reference image.

[0053] <Embodiment 3> Fig. 12 is a diagram showing another example of the sample 40 and its image feature amounts. Using Fig. 12, an example of the relationship between the type of the sample 40 and the image feature amounts will be described.

[0054] The left side of FIG. 12 is an example of an observation image when the immunochromatography test kit described in Embodiment 1 is used as Sample 40. This example shows a case where the numerical values of the feature amounts at the inspection position 42 and the reference position 43 are small respectively.

[0055] The right side of FIG. 12 is an example of an observation image when fibers on a flat plate are used as Sample 40. That is, an example is shown in which the inspection object 125 included in the observation image is a fiber. In this example, the purpose of the inspection is to count the number of fibers. The image feature amounts are the number of fibers, the area, etc. In this example, the fibers are scattered over the entire flat plate and the number of fibers is large. Therefore, the numerical values of the feature amounts are large at both the inspection position and the reference position. Even in this case, a statistically significant difference may occur between the image feature amounts at each position. Therefore, it is possible to determine whether there is a significant difference between the image feature amount at the inspection position and the image feature amount at the reference position by using the methods described in Embodiments 1 to 2. That is, if a significant difference occurs between the image feature amounts at each position, the type of Sample 40 is arbitrary.

[0056] <Regarding the modification of the present invention> The present invention is not limited to the above-described embodiments and includes various modifications. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Also, a part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can also be added to the configuration of one embodiment. Further, it is possible to add, delete, or replace other configurations for a part of the configuration of each embodiment.

[0057] In the above embodiments, the user may specify a specific position on Sample 40 on the user interface provided by the input / output device 30 and acquire an observation image at that position. In this case, the selection unit 32 receives the specified position and instructs the image acquisition control unit 11 to acquire an observation image at the specified position.

[0058] In the above embodiments, an example in which the image acquisition device 10, the evaluation device 20, and the input / output device 30 are configured as part of the inspection device 1 has been described. However, these may be configured as separate devices from each other. For example, by connecting each device via a network, it is possible to configure each device as an individual device and construct the same configuration as in the above embodiments.

[0059] In the above embodiments, at least any one of the functional units included in the evaluation device 20 can also be integrally configured. For example, the reliability evaluation unit 26 may be configured to include the feature quantity comparator 25 and the occupancy rate analysis unit 24. The same applies to other functional units.

[0060] In the above embodiments, it is also possible to construct the inspection device 1 according to the present invention by installing software that implements the functions of the evaluation device 20 according to the present invention in an existing inspection device 1.

[0061] In the above embodiments, examples of the inspection target by the inspection device 1 include antigens, antibodies, etc. containing viruses or bacteria. When using an immunochromatography test kit for testing antigens and antibodies as the sample 40, components of the sample 40 include particles, voids, a substrate, a liquid, etc. The components of the sample 40 are not limited to this.

Explanation of Reference Numerals

[0062] 1: Inspection device 10: Image acquisition device 20: Evaluation device 21: Image correction unit 22: Feature quantity extraction unit 23: Target area specification unit 24: Occupancy rate analysis unit 25: Feature quantity comparator 26: Reliability evaluation unit 27: Reacquisition determination unit 28: Storage unit 30: Input / output device

Claims

1. An inspection apparatus for inspecting a sample using an observation image of the sample, comprising: a feature quantity extraction unit that extracts a feature quantity of the observation image; a reliability evaluation unit that evaluates the reliability of the feature quantity; and the inspection apparatus further comprises a reacquisition determination unit that determines whether it is necessary to reacquire the observation image after the reliability evaluation unit evaluates the reliability; when the reliability is not greater than or equal to a reacquisition threshold, the reacquisition determination unit determines that it is necessary to reacquire the observation image; when it is determined that it is necessary to reacquire the observation image, the feature quantity extraction unit re-extracts the feature quantity of the reacquired observation image; when it is determined that it is necessary to reacquire the observation image, the reliability evaluation unit re-evaluates the reliability of the re-extracted feature quantity; even when the reliability is not greater than or equal to the reacquisition threshold, if the number of the reacquired observation images reaches a predetermined number or more, the reacquisition determination unit determines that it is not necessary to reacquire the observation image An inspection apparatus characterized by the above.

2. An inspection apparatus for inspecting a sample using an observation image of the sample, comprising: a feature quantity extraction unit that extracts a feature quantity of the observation image; a reliability evaluation unit that evaluates the reliability of the feature quantity; and the feature quantity extraction unit extracts a first feature quantity from a target region including an inspection target in the observation image; the feature quantity extraction unit extracts a second feature quantity from a reference region other than the target region in the observation image; the reliability evaluation unit calculates the reliability of the first feature quantity by comparing the first feature quantity and the second feature quantity; the inspection apparatus further comprises a reacquisition determination unit that determines whether it is necessary to reacquire the observation image after the reliability evaluation unit evaluates the reliability; when the reliability is not greater than or equal to a reacquisition threshold, the reacquisition determination unit determines that it is necessary to reacquire the observation image; when it is determined that it is necessary to reacquire the observation image, the feature quantity extraction unit re-extracts the feature quantity of the reacquired observation image; when it is determined that it is necessary to reacquire the observation image, the reliability evaluation unit re-evaluates the reliability of the re-extracted feature quantity; when the reason that the reliability is not greater than or equal to the reacquisition threshold is due to insufficient amount of the observation image in the reference region, the reacquisition determination unit determines that it is necessary to reacquire the observation image in the reference region When it is determined that it is necessary to reacquire the observation image in the reference region, the feature amount extraction unit re - extracts the second feature amount from the observation image in the reacquired reference region, When it is determined that it is necessary to reacquire the observation image in the reference region, the reliability evaluation unit calculates the reliability by comparing the re - extracted second feature amount with the first feature amount An inspection apparatus characterized by the above.

3. An inspection apparatus for inspecting a sample using an observation image of the sample, A feature amount extraction unit that extracts feature amounts of the observation image, A reliability evaluation unit that evaluates the reliability of the feature amounts, Comprising: The inspection apparatus further comprises a target region specifying unit that specifies a target region including an inspection target from the observation image, The target region specifying unit determines a portion of the observation image that is the inspection target and a portion that is not, using at least one of the feature amounts of size, shape, and luminance value, The target region specifying unit specifies, as the target region, a region excluding a portion of the observation image determined not to be the inspection target. An inspection apparatus characterized by the above.

4. The feature amount extraction unit extracts a first feature amount from a target region including an inspection target in the observation image, The feature amount extraction unit extracts a second feature amount from a reference region other than the target region in the observation image, The reliability evaluation unit calculates the reliability of the first feature amount by comparing the first feature amount with the second feature amount. The inspection apparatus according to claim 1 or 3, characterized by the above.

5. The feature amount extraction unit extracts a first feature amount from a target region including an inspection target in the observation image, The feature amount extraction unit extracts a second feature amount from a reference region other than the target region in the observation image, The reliability evaluation unit calculates the reliability of the first feature amount by comparing the first feature amount with the second feature amount, The feature amount extraction unit calculates, as the first feature amount, a first statistic of pixel values of the observation image in the target region, The feature amount extraction unit calculates, as the second feature amount, a second statistic of pixel values of the observation image in the reference region, The reliability evaluation unit calculates an evaluation value indicating whether there is a significant difference between the first feature amount and the second feature amount by comparing the first statistic with the second statistic, The reliability evaluation unit calculates the reliability using the evaluation value The inspection apparatus according to claim 1 or 3, characterized in that...

6. The feature quantity extraction unit calculates, as the first feature quantity, a first statistic of the pixel values of the observation image in the target region. The feature quantity extraction unit calculates, as the second feature quantity, a second statistic of the pixel values of the observation image in the reference region. The reliability evaluation unit calculates an evaluation value indicating whether there is a significant difference between the first feature quantity and the second feature quantity by comparing the first statistic and the second statistic. The reliability evaluation unit calculates the reliability using the evaluation value. The inspection apparatus according to claim 2, characterized in that...

7. The feature quantity extraction unit extracts a first feature quantity from a target region including an inspection target in the observation image. The feature quantity extraction unit extracts a second feature quantity from a reference region other than the target region in the observation image. The reliability evaluation unit calculates the reliability of the first feature quantity by comparing the first feature quantity and the second feature quantity. The reliability evaluation unit calculates the reliability such that the greater the significant difference between the first feature quantity and the second feature quantity, the higher the reliability. The inspection apparatus according to claim 1 or 3, characterized in that...

8. The reliability evaluation unit calculates the reliability such that the greater the significant difference between the first feature quantity and the second feature quantity, the higher the reliability. The inspection apparatus according to claim 2, characterized in that...

9. When the reason that the reliability is not greater than the reacquisition threshold is due to insufficient amount of the observation image in the reference region, the reacquisition determination unit reacquires the observation image in the reference region and then reacquires the observation image in the target region. The inspection apparatus according to claim 2, characterized in that...

10. The feature quantity extraction unit extracts, as a derived feature quantity of the observation image, a parameter representing the ratio of the pixels having the feature quantity in the target region to the area of the target region among the pixels of the observation image in the target region. The inspection apparatus according to claim 3, characterized in that...

11. The feature quantity extraction unit extracts a first feature quantity from a target region including an inspection target in the observation image. The feature quantity extraction unit extracts a second feature quantity from a reference region other than the target region in the observation image. The reliability evaluation unit calculates the reliability of the first feature quantity by comparing the first feature quantity and the second feature quantity. The feature extraction unit extracts, as a derived feature amount of the observation image, a parameter representing a ratio of the pixels of the observation image in the target area having the feature amount to the area of the target area. The reliability evaluation unit calculates the reliability of the first feature amount by comparing the derived feature amount in the target area with the derived feature amount in the reference area. The inspection apparatus according to claim 1 or 3, characterized in that.

12. The feature extraction unit extracts a first feature amount from a target area including an inspection target in the observation image. The feature extraction unit extracts a second feature amount from a reference area other than the target area in the observation image. The reliability evaluation unit calculates the reliability of the first feature amount by comparing the first feature amount with the second feature amount. The inspection apparatus further includes a correction unit that corrects at least one of the observation image in the target area or the observation image in the reference area. The inspection apparatus according to claim 1 or 3, characterized in that.

13. The inspection apparatus further includes a correction unit that corrects at least one of the observation image in the target area or the observation image in the reference area. The inspection apparatus according to claim 2, characterized in that.

14. The sample has at least one of an antigen or an antibody as an object of the inspection. The inspection apparatus according to any one of claims 1 to 3, characterized in that.

15. The feature extraction unit extracts the number of marker particles attached to the antigen or antibody as a feature amount of the observation image. The inspection apparatus according to claim 14, characterized in that.

16. The inspection apparatus is configured as an apparatus that acquires the observation image by a charged particle beam. The inspection apparatus according to any one of claims 1 to 3, characterized in that.

17. The sample contains at least one of particles, voids, a substrate, and a liquid. The inspection apparatus according to any one of claims 1 to 3, characterized in that.

Images