JP2010217104A - Image analyzer, image analyzing method and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image analyzer capable of efficiently extracting a specific image from a plurality of images, and an image analyzing method and a program therefor. <P>SOLUTION: The image analyzing method, for obtaining a plurality of time series images to analyze and operate, includes the steps of: checking the plurality of images in a chronological order to detect a first image satisfying a first requirement; checking an image following the first image to detect a second image not satisfying the first requirement; extracting a series of images from the first image to the second image; and analyzing and operating by using a new plurality of images excluding the series of images from the plurality of images, in which the first requirement is that there is a region with an intensity value greater than a first intensity value in an image. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an image analysis technique for extracting a specific image from a plurality of images.

  RICS (Raster-scan Image Correlation Spectroscopy) is a technique for calculating the diffusion rate and concentration of molecules by spatial correlation calculation of signal intensity for each image pixel based on a scan image such as a microscope image.

  Non-Patent Document 1 describes the RICS developer Gratton et al. Focusing on the explanation of the principle of RICS, and also describes the measurement in cells, and the localization of GFP-Paxillin relative to GFP-expressing cells. The example which measured diffusion by RICS is given. Paxillin is a protein that interacts with cell membrane proteins such as actin and integrins inside the cell membrane and is related to adhesion to the extracellular matrix. In this paper, the results of RICS measurement of localized Paxillin-GFP are shown, and the diffusion rate almost 10 times slower than that of GFP alone was obtained, which appeals to the practicality of RICS measurement.

  Non-Patent Document 2 is a detailed investigation of image acquisition conditions and correlation function calculation conditions for RICS. This suggests that an image used for analysis requires a certain number of images to increase the S / N.

MA.Digman et.al., “Measuring Fast Dynamics in Solutions and Cells with a Laser Scanning Microscope.”, Biophysical Journal Vol. 89, 1317-1327, 2005. CM. Brown et al. “Raster image correlation spectroscopy (RICS) for measuring fast protein dynamics and concentrations with a commercial laser scanning confocal microscope.” J Microsc. 229, 78-91. 2008.

  By the way, when statistically measuring the movement of an object (for example, a molecule), a target whose size and diffusion rate are uniform or whose molecular size increases due to physiological intermolecular interactions, etc., resulting in slow diffusion. Examine the behavior of molecules. At this time, it is feared that unexpected abnormal molecules (such as aggregates) flow into ROI (Region of interest) with high luminance during the measurement, thereby reducing the reliability of the final result.

  In RICS measurement, calculation is performed by acquiring a plurality of image frames, so that high-intensity fluorescent molecules (such as aggregated molecules) that pass through the observation region during image acquisition can be removed for each image frame. However, in order to remove this problematic frame from about 100 consecutive frame images, it is necessary to check the details one by one, which is a cumbersome and time-consuming operation.

  This invention is made | formed in view of the situation which concerns, Comprising: It aims at providing the image-analysis apparatus, the image-analysis method, and program which can extract a specific image efficiently from several images.

  In order to solve the above-described problems, the present invention provides an image analysis apparatus that acquires a plurality of time-series images and performs an analysis operation. The plurality of images are examined in time-series order and the first condition is satisfied first. A first image detecting means for detecting the first image to be detected, and a second image detecting the second image not satisfying the first condition first by examining images following the first image in time series. An image detecting means; a first image extracting means for extracting a series of images from the first image to the second image; and a plurality of new images obtained by removing the series of images from the plurality of images. And an analysis unit that executes an analysis operation using an image, wherein the first condition is an image analysis apparatus in which a region having a first luminance value or more exists in the image.

  In the image analysis method for obtaining a plurality of time-series images and performing an analysis operation, the present invention examines the plurality of images in time-series order, and firstly satisfies the first condition. A second image that does not satisfy the first condition is detected by examining images following the first image in time series order, and a series of steps from the first image to the second image are detected. An image is extracted, and an analysis operation is performed using a plurality of new images obtained by removing the series of images from the plurality of images. The first condition is equal to or greater than a first luminance value in the images. This is an image analysis method in which the region exists.

  According to the present invention, in an image analysis program for acquiring a plurality of time-series images and performing an analysis operation, the plurality of images are examined in time-series order, and a first image that satisfies the first condition first is obtained. A procedure for detecting, a procedure for examining images following the first image in chronological order, and detecting a second image that does not satisfy the first condition first, and from the first image to the second image A procedure for extracting a series of images, causing a computer to execute a procedure for performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images, wherein the first condition includes: This is a program in which an area having a first luminance value or more exists in an image.

  According to the present invention, it is possible to provide an image analysis apparatus, an image analysis method, and a program that can efficiently extract a specific image from a plurality of images.

The figure which shows the structure of the laser microscope system to which the image-analysis method of this Embodiment is applied. The schematic diagram which shows the various quantities relevant to a measurement with a sufficient precision. The figure explaining the acquisition operation | movement of a flame | frame. The figure for demonstrating the content of a spatial correlation calculation. The figure which shows a 2D correlation function figure. The figure for demonstrating the fitting method. The figure which shows the rough process sequence which extracts the flame | frame containing a high-intensity molecule | numerator. The figure which shows the rough process sequence which extracts the flame | frame containing a high-intensity molecule | numerator. The figure which shows typically the process which extracts the flame | frame containing a high-intensity molecule | numerator.

FIG. 1 is a diagram showing a configuration of a laser microscope system to which the image analysis method of the present embodiment is applied.
The laser microscope system includes a microscope main body 1, a laser combiner 2, a scan unit 3, a detector unit 4, a control unit 5, a display device 6 and a storage device 7.

  Laser light emitted from one laser light source provided in the laser combiner 2 is reflected by a dichroic mirror (not shown) and then enters the scan unit 3. The scanning unit 3 deflects and scans the optical axis in the X-axis and Y-axis directions and enters an objective lens (not shown) of the microscope body 1. As a result, the measurement region (focal region) can be positioned at an arbitrary position of the fluorescently labeled sample specimen on the focal plane in the field of view.

  Fluorescence emitted from the fluorescent molecules in the focal region is captured by the same objective lens and guided to the dichroic mirror through the reverse optical path. The dichroic mirror is designed to transmit fluorescence having a longer wavelength than the excitation light, and the fluorescence reaches the detector unit 4. The detector unit 4 is preferably an APD (avalanche photodiode) or a photomultiplier tube.

  The fluorescence intensity signal photoelectrically converted by the detector unit 4 is input to the control unit 5. In the control unit 5, RICS analysis is performed by analysis software. The control unit 5 performs RICS analysis by associating the fluorescence intensity signal from the detector unit 4 with the scan position information from the scan unit 3. The result of the RICS analysis is appropriately displayed on the display device 6 and stored in the storage device 7.

  In this laser microscope system, a plurality of fluorescent images having different wavelengths can be obtained using a plurality of laser light sources. In the following description, an analysis is performed using a fluorescent image obtained by one laser light source. explain.

On the right side of FIG. 1, information obtained by the laser microscope system is schematically shown.
The sample 8a is repeatedly scanned and irradiated with laser light to obtain a plurality of fluorescent images 8b. Based on the respective fluorescence images 8b, data 8c relating to the fluorescence intensity is obtained and input to the analysis software. The analysis software performs RICS analysis based on the respective data 8c. That is, spatial correlation calculation is performed to obtain a spatial correlation function diagram 8d, and a fitting analysis is performed between the result and the reference spatial correlation function diagram.

Next, the contents of RICS analysis will be described in detail.
* Step 1: Scan with laser light so that target (fluorescent) molecules can be detected pixel by pixel.
FIG. 2 is a schematic diagram showing various quantities related to accurate measurement.
FIG. 2 shows the moving speed (trajectory) of the target 10, the size of the laser spot 11 irradiated on the target 10, and the pixel position (1, 2,...) Of the detection element.

  According to FIG. 2, it can be seen that there is a close relationship among the moving speed of the target, the scanning speed, and the pixel size (actual size per pixel). Therefore, when the number of molecules of the target, the diffusion rate, and the like are obtained by RICS analysis, it is important that various quantities shown in FIG. 2 are set to appropriate values.

* Step 2: Scan the laser within the set frame. This scan is executed a plurality of times to obtain a plurality of frames.
FIG. 3 is a diagram for explaining a frame acquisition operation.
The size of one frame 12 (vertical length × horizontal length) is set by the analyst in step 1. The laser beam is scanned by the scan unit 3 in this frame area. For example, the laser beam is scanned one line from the upper left position of the frame area to the right, and then the lower line is scanned from the left to the right. This operation is repeated up to the bottom line to complete one frame scan.
This frame scanning is repeated and executed a plurality of times in the same manner to obtain a plurality of frame images.

  Here, the time for scanning one pixel is on the order of μsec, the time for scanning one line is on the order of msec, and the time for scanning one frame is on the order of sec. When one frame scanning time is 1 second, for example, and 100 frames are acquired, the measurement time is about 100 seconds.

  In principle, one frame screen is sufficient for performing spatial correlation calculations. However, in this embodiment, the average calculation regarding the spatial correlation calculation is performed by increasing the number of acquired frames. By increasing the number of frames and calculating the average value of the correlation function, it is possible to remove molecular information of slow motion.

* Step 3: Perform spatial correlation calculation.
The spatial correlation calculation is executed according to the following correlation function expression.

FIG. 4 is a diagram for explaining the contents of the spatial correlation calculation.
Two identical frames 12 are superposed with a shift of ξ in the X-axis direction and ψ in the Y-axis direction. Then, a value obtained by multiplying the fluorescence intensity values at the overlapping positions is added over the entire frame. A value obtained by normalizing the addition result is set as a spatial correlation value at the point (ξ, ψ). By changing these ξ and ψ on a two-dimensional plane, a two-dimensional spatial correlation value is obtained.

  Next, a two-dimensional spatial correlation value is calculated for each of the other frames. The plurality of spatial correlation values acquired in this way are averaged for each point to obtain a 2D correlation function diagram (spatial correlation function diagram).

FIG. 5 is a diagram showing a 2D correlation function diagram.
As can be seen from the above formula, the correlation function value takes the maximum value (= 1) at the origin (0, 0) located at the center of FIG.

* Step 4: A fitting calculation is performed using a 2D correlation function diagram obtained as a result of the correlation function calculation.
When the fluorescence intensity values are plotted in the height direction (Z-axis direction) based on the 2D correlation function diagram obtained in this manner, a mountain shape centering on the origin is obtained. The measured shape is compared with the shapes of a plurality of reference patterns to identify a reference pattern that fits well.

FIG. 6 is a diagram for explaining the fitting method.
On the lower side of the figure, the correlation function values obtained by the calculation are represented in three dimensions. On the upper side of the figure, the difference from a certain reference pattern is shown in three dimensions. A reference pattern that is evaluated to have the smallest difference is specified. Then, the number of molecules and the diffusion rate corresponding to the specified reference pattern can be obtained as the average number of molecules and the average diffusion rate of the sample.

By the way, in the RICS calculation described above, a process for excluding a frame in which a disturbance factor exists is required so as not to be affected by the disturbance factor during measurement.
That is, as described above, it is feared that unexpected molecules (aggregates and the like) flow into ROI (Region of interest) with high luminance during measurement, thereby reducing the reliability of the final result. Therefore, it is desired that a frame including such high-luminance molecules can be extracted efficiently.

Here, the basic concept in the present invention will be described based on the RICS analysis example performed by the inventor.
The analysis sample used was a fluorescent (protein) -labeled protein observed in a solution or in a cell, and movement information of the molecule and size information derived therefrom were measured by a technique such as RICS.

  For labeling, fluorescent molecules (rhodamine, cy3, cy5, Alexa-, Atto-, etc.) or fluorescent proteins (EGFP, mDsRed, mRFP, mCherry, YFP, etc.) were used. In addition, depending on the purpose, the interaction between two or more proteins labeled with both was also measured by observing two colors of fluorescence simultaneously.

In this series of measurements, the above-described molecules were observed in the following manner.
Case 1: Removal of intracellular protein aggregation
In EGFP-only cells, EGFP normally has no interaction with other molecules and cell organs, but may aggregate due to the environment (intracellular pH, temperature, etc.) where EGFP is present. . At that time, other molecules may be incorporated to form giant aggregates.

  For these reasons, there are cases where aggregated high-brightness macromolecules are detected when image frames are continuously acquired. When measuring intracellular kinetics of EGFP or EGFP fusion protein, it is necessary to exclude these high brightness molecules and perform correlation calculation.

Case 2: Detection of intracellular aggregated protein
PolyQ aggregation (continuous binding protein of glutamate) causes neuronal cell death and is thought to cause Huntington's disease. Observing and measuring the progress of intracellular polyQ protein from monomer to polymer and large aggregation is important for exploring its aggregation mechanism and disease cause.

  For example, a fluorescent protein such as EGFP is fused with PolyQ and expressed in cells. When PolyQ-EGFP grows, the size increases and the fluorescence intensity also increases. In the monomer state, PolyQ-EGFP diffuses freely in the cell, but as aggregation increases, it changes to a large, bright molecule, thereby slowing the molecular diffusion rate.

  When this reaction is observed by an image measurement method such as RICS, when a plurality of images are acquired, high-intensity molecules flow into the frame as aggregated proteins, so that their presence and size can be estimated.

The following is derived from this result.
(1) High-brightness molecules observed in the RICS analysis should be excluded as disturbance factors, but the behavior of the high-brightness molecules themselves includes those that can be separately analyzed.
(2) The high-brightness molecules observed in the RICS analysis include a state in which the luminance is high from the beginning and a state in which the luminance and size are gradually condensed to increase the luminance and size.

The image analysis method of the present embodiment based on the above examination results will be described.
7 and 8 are diagrams showing a schematic processing procedure for extracting a frame including high-luminance molecules.

  In step S01, the analyst determines the magnification of the sample and a region to be photographed in the observation field by observing with a microscope. In step S02, a region (ROI) to be irradiated with laser light is set. In step S03, the number of frames to be acquired is set.

  Then, when an analysis start is input to the system, a frame acquisition process is executed in step S04. That is, the scanning unit 3 scans the region (ROI) with laser light, and the number of frames in which a frame representing a fluorescent image of the sample specimen is set is acquired. The acquired frames are collected into one file and stored in the storage device 7.

Subsequently, an extraction process is performed on frames containing high-luminance molecules.
In step S09, initialization of each parameter variable is executed. That is, page i = 1, first start page SP1 = first last page LP1 = second start page SP2 = second last page LP2 = 0.

In step S10, the frame of the i-th page is extracted from the file in the storage device 7. In step S11, it is checked whether or not there is a region having a fluorescence intensity of a predetermined value (F1) or more in the frame.
If Yes in step S11, that is, if there is a region having a fluorescence intensity greater than or equal to the predetermined value (F1) in the frame, whether or not a value of 1 or greater is set in the first start page SP1 in step S12. Investigate.

In the case of No in step S12, that is, when the first start page SP1 is the initial value (= 0), an area having a fluorescence intensity greater than or equal to a predetermined value (F1) is detected for the first time in this i-th frame. , The first start page SP1 = i is set. Then, the process of step S21 is executed.
In the case of Yes in step S12, that is, when a value of 1 or more is set in the first start page SP1, a region having a fluorescence intensity of the predetermined value (F1) or more has already been detected in another frame. , Do no special processing. Then, the process of step S21 is executed.

  If No in step S11, that is, if there is no region having a fluorescence intensity greater than or equal to the predetermined value (F1) in the frame, whether or not a value of 1 or greater is set in the first start page SP1 in step S14. Investigate.

If No in step S14, that is, if the first start page SP1 is the initial value (= 0), a region having a fluorescence intensity greater than or equal to the predetermined value (F1) has not yet been detected, and special processing is performed. Absent. Then, the process of step S21 is executed.
In the case of Yes in step S14, that is, when a value of 1 or more is set in the first start page SP1, a region having a fluorescence intensity of the predetermined value (F1) or more has already been detected in another frame. In this i-th frame, it was detected that a region having a fluorescence intensity equal to or higher than a predetermined value (F1) disappeared (flowed out of the frame). Therefore, in step S15, the first last page LP1 = i is set. Then, the process of step S31 is executed.

In step S21 of FIG. 8, it is checked whether or not there is a region having a fluorescence intensity of a predetermined value F2 (<F1) or more and an area of S or more in the frame.
In the case of Yes in step S21, that is, in the case where there is a region having a fluorescence intensity greater than or equal to a predetermined value F2 (<F1) in the frame and the area of the region is greater than or equal to S, in step S22, the second start page SP2 It is checked whether or not a value of 1 or more is set.

In the case of No in step S22, that is, when the second start page SP2 is the initial value (= 0), this i-th frame has a fluorescence intensity equal to or higher than a predetermined value F2 (<F1) for the first time, and Since the area having an area of S or more is detected, the second start page SP2 = i is set in step S23. Then, the process of step S26 is executed.
In the case of Yes in step S22, that is, when a value of 1 or more is set in the second start page SP2, it has a fluorescence intensity of a predetermined value F2 (<F1) or more and the area of the region is S or more. Since the area has already been detected in another frame, no special processing is performed. Then, the process of step S26 is executed.

  In step S26, page i is incremented by one. In step S27, it is checked whether or not all frames have been processed. If No in step S27, that is, if processing has not been performed for all frames, the processing returns to step S10 and the above-described processing is repeated. If Yes in step S27, that is, if the processing is completed for all frames, the processing in step S31 is executed.

  In the case of No in step S21, that is, in the case where there is no region having a fluorescence intensity greater than or equal to a predetermined value F2 (<F1) and the area of the region is greater than or equal to S, in step S24, the second start page SP2 It is checked whether or not a value of 1 or more is set.

  In the case of No in step S24, that is, when the second start page SP2 is an initial value (= 0), a region having a fluorescence intensity of a predetermined value F2 (<F1) or more and an area of the region of S or more is Since it has not been detected yet, no special processing is performed. Then, the process of step S26 is executed.

  In step S26, page i is incremented by one. In step S27, it is checked whether or not all frames have been processed. If No in step S27, that is, if processing has not been performed for all frames, the processing returns to step S10 and the above-described processing is repeated. If Yes in step S27, that is, if the processing is completed for all frames, the processing in step S31 is executed.

  In the case of Yes in step S24, that is, when a value of 1 or more is set in the second start page SP2, it has a fluorescence intensity of a predetermined value F2 (<F1) or more and the area of the region is S or more. Since the area has already been detected in another frame, it is considered that it has been detected that the area has disappeared (flowed out of the frame) in this i-th frame. Therefore, in step S25, the second last page LP2 = i is set. Then, the process of step S31 is executed.

Next, frame exclusion processing is executed based on the extracted result.
In step S31, it is checked whether or not a value of 1 or more is set in SP1 or SP2. If Yes in step S31, that is, if a value of 1 or more is set in SP1 or SP2, in step S32, a new file obtained by removing pages SP1 to LP1 or pages SP2 to LP2 from the recorded file. A file is generated and stored in the storage device 7.
Although a value of 1 or more is set in SP1 or SP2, if the corresponding LP1 or LP2 is 0, the last page of the file is set in LP1 or LP2.

  In step S33, a frame corresponding to pages SP1 to LP1 or pages SP2 to LP2 is generated as another file and stored in the storage device 7. In step S40, a new processed file is read from the storage device 7 and RICS analysis is executed.

  In the case of No in step S31, that is, when SP1 and SP2 are in the initial value state (= 0), no high-luminance numerator exists, so in step S34, no processing is performed on the frame. In step S40, a new processed file is read from the storage device 7 and RICS analysis is executed.

FIG. 9 is a diagram schematically illustrating processing for extracting a frame including high-luminance molecules.
When there are n frames in all in one file, high-brightness molecules are observed in the i-th frame, and when no high-brightness molecules are observed in the o-th frame, While the i-th to o-th frames are removed, another file is configured with the i-th to o-th frames.

  In addition, when confirming the frame to be removed, it may be processed fully automatically as described above, or a plurality of frames including the extracted i-th to o-th frames are displayed on the display device 6, and based on that. The frame selected by the analyst may be removed.

  According to the embodiment described above, various effects can be achieved.

(1) Two different criteria are provided as conditions for determining the presence of high-luminance molecules.
One criterion is a case where a region with high luminance (F1 or higher) exists, and this is a condition for detecting that a high luminance macromolecule with high luminance has flowed into the frame from the beginning. In this case, the luminance value can be set to, for example, about 100 times or more of the usual fluorescence intensity (of a sample that is not a high luminance molecule).

Another criterion is a case where there is a region where the area of high brightness F2 (<F1) is S or more, and this is a condition for detecting that the size increases and the fluorescence intensity increases at the stage of molecular growth. It is. The luminance value in this case can be set to, for example, about 50 times or more of the usual fluorescence intensity (of a sample that is not a high luminance molecule). For example, the size can be set to 10 or more in terms of the number of pixels.
The above values of 100 times, 50 times, and 10 are values extracted based on the results observed by the inventor on the actual machine. Therefore, by processing according to this value, it is possible to automatically determine what type of high brightness molecule exists.

(2) The detection result can be used effectively.
The frame in which the detected high-brightness molecule exists is not only removed from the original file, but also created as a separate file. As a result, the newly created file can be used for high brightness molecular analysis.
In the above-described processing flow, all are extracted and stored regardless of the detection criterion, but only for the case where there is a region where the area of high luminance F2 (<F1) is S or more, for high luminance molecular analysis. May be saved.

(3) It can process efficiently.

  According to the above processing method, it is possible to easily and efficiently determine the presence or absence of high-luminance molecules by examining only the luminance value and size in the frame.

Note that the present invention is not limited to the above-described embodiment as it is, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage.
Further, various inventions can be formed by appropriately combining a plurality of constituent elements disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, you may combine suitably the component covering different embodiment.

    DESCRIPTION OF SYMBOLS 1 ... Microscope main body, 2 ... Laser combiner, 3 ... Scan unit, 4 ... Detector unit, 5 ... Control unit, 6 ... Display apparatus, 7 ... Memory | storage device, 8a ... Sample, 8b ... Fluorescent image, 12 ... Frame.

Claims (9)

In an image analysis apparatus that acquires a plurality of time-series images and performs an analysis operation,
First image detecting means for examining the plurality of images in chronological order and detecting a first image that first satisfies a first condition;
Second image detection means for examining images following the first image in time series order and detecting a second image that does not satisfy the first condition first;
First image extraction means for extracting a series of images from the first image to the second image;
Analysis means for performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images,
The image analysis apparatus according to claim 1, wherein the first condition is that an area having a luminance value equal to or higher than the first luminance value exists in the image. A third image detection means for examining the plurality of images in chronological order and detecting a third image that satisfies a second condition first;
A fourth image detecting means for examining images following the third image in time-series order and first detecting a fourth image not satisfying the second condition;
A second image extracting means for extracting a series of images from the third image to the fourth image;
The analysis means performs an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images.
The second condition is characterized in that a region having a second luminance value smaller than the first luminance value is present in the image and the area is equal to or larger than a predetermined value. The image analysis apparatus described in 1.   The image analysis apparatus according to claim 1, further comprising another image storage unit that stores the extracted series of images as another plurality of time-series images.   The image analysis apparatus according to claim 3, wherein the plurality of time-series images are time-series images obtained by optically scanning the same measurement visual field.   The image analysis apparatus according to claim 4, wherein the analysis calculation is a RICS analysis including a spatial correlation calculation. In an image analysis method for acquiring a plurality of time-series images and performing an analysis operation,
Examining the plurality of images in chronological order to detect a first image that satisfies the first condition first,
The images following the first image are examined in chronological order to first detect a second image that does not satisfy the first condition,
Extracting a series of images from the first image to the second image;
Performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images,
The image analysis method according to claim 1, wherein the first condition is that an area having a luminance value equal to or higher than a first luminance value exists in the image. Examining the plurality of images in chronological order to detect a third image that satisfies the second condition first,
The images following the third image are examined in chronological order to detect a fourth image that does not satisfy the second condition first,
Extracting a series of images from the third image to the fourth image;
Performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images,
8. The second condition is characterized in that a region having a second luminance value that is smaller than the first luminance value is present in the image and the area of the region is a predetermined value or more. The image analysis method described in 1. In an image analysis program that acquires multiple time-series images and performs analysis operations,
A procedure for examining the plurality of images in chronological order and detecting a first image satisfying a first condition first;
A procedure for detecting a second image that does not satisfy the first condition by examining images following the first image in time series order,
Extracting a series of images from the first image to the second image;
Causing the computer to execute a procedure of performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images;
The first condition is a program characterized in that an area having a first luminance value or more exists in the image. A procedure for examining the plurality of images in chronological order and first detecting a third image satisfying the second condition;
A procedure for examining images following the third image in chronological order and first detecting a fourth image not satisfying the second condition;
A procedure for extracting a series of images from the third image to the fourth image;
A procedure for performing an analysis operation using a plurality of new images obtained by removing the series of images from the plurality of images;
Is further executed on the computer,
9. The second condition is characterized in that an area having a second luminance value smaller than the first luminance value is present in the image and the area is equal to or larger than a predetermined value. The program described in.

Images