Observation equipment and observation method

Abstract

The observation device is configured to include a threshold set acquisition unit (2) that acquires N (N is an integer of 2 or more) different threshold sets as threshold sets including thresholds for one or more types of observation components included in both a first observation image, which is an image of the observation target area captured, and a second observation image, which is an image of the observation target area captured at a time different from that of the first observation image, and a difference calculation unit (3) that calculates a difference between each observation component included in the first observation image and each observation component included in the second observation image. The observation device also includes a comparison unit (4) that compares the difference of each observation component calculated by the difference calculation unit (3) with a threshold for each observation component included in each threshold set acquired by the threshold set acquisition unit (2), and a threshold set selection unit (5) that selects one of the threshold sets acquired by the threshold set acquisition unit (2) based on a result of the comparison by the comparison unit (4).

Description

[Technical field]

[0001] The present disclosure relates to an observation device and an observation method. [Background technology]

[0002] There is an observation device that calculates the difference between multiple observation images of an observation target area taken at different times, compares the difference with a threshold, and detects changes in the observation target area based on the comparison result between the difference and the threshold. As an example of such an observation device, Patent Document 1 discloses an observation device having a threshold set including thresholds for each of one or more types of observation components contained in an observed image, and equipped with a comparison unit that compares each observation component with the threshold for that observation component. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2003-281664 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the observation device disclosed in Patent Document 1, the comparison unit has only one threshold set. Therefore, the thresholds for each observation component included in the threshold set may not be suitable for comparison with the difference. In such a case, there is a problem that the detection accuracy of the change in the observation target area is deteriorated.

[0005] The present disclosure has been made to solve the above-mentioned problems, and aims to obtain an observation device that can detect changes in the observation target area with higher accuracy than the observation device disclosed in Patent Document 1. [Means for solving the problem]

[0006] The observation device according to the present disclosure includes a threshold set acquisition unit that acquires N (N is an integer equal to or greater than 2) different threshold sets as threshold sets including thresholds for one or more types of observation components included in both a first observation image, which is an image of an observation target area captured, and a second observation image, which is an image of the observation target area captured at a time different from that of the first observation image, and a difference calculation unit that calculates a difference between each observation component included in the first observation image and each observation component included in the second observation image. The observation device also includes a comparison unit that compares the difference between each observation component calculated by the difference calculation unit with a threshold for each observation component included in each threshold set acquired by the threshold set acquisition unit, and a threshold set selection unit that selects one of the threshold sets acquired by the threshold set acquisition unit based on a result of the comparison by the comparison unit. Effect of the Invention

[0007] According to the present disclosure, it is possible to improve the accuracy of detecting changes in the observation target area compared to the observation device disclosed in Patent Document 1. [Brief description of the drawings]

[0008] [Figure 1] 1 is a configuration diagram showing an observation device according to a first embodiment. [Diagram 2] 2 is a hardware configuration diagram showing hardware of an observation device according to the first embodiment. FIG. [Diagram 3] FIG. 2 is a hardware configuration diagram of a computer in the case where the observation device is realized by software, firmware, or the like. [Figure 4] 4 is a flowchart showing an observation method which is a processing procedure of the observation device. [Diagram 5] 1A and 1B are explanatory diagrams showing a first observation image G1 and an observation target image G2,m, respectively. [Figure 6] FIG. 11 is an explanatory diagram showing an example of a weighting coefficient Wd u(td u) that changes with time. [Figure 7]FIG. 13 is an explanatory diagram showing an example of an area for which importance is set by a user. [Figure 8] 11 is an explanatory diagram showing an example of local information indicating whether or not an abnormality has occurred in an observation target area; FIG. [Figure 9] FIG. 9 is an explanatory diagram in which FIG. 7 and FIG. 8 are superimposed. [Figure 10] FIG. 11 is an explanatory diagram showing N threshold sets S1 to SN, a comparison result Rm,jn related to a difference ΔOCm,j of an observed component, a maximum score Ωm,MAX, and a threshold set Sm,MAX. [Figure 11] FIG. 11 is a configuration diagram showing an observation device according to a second embodiment. [Figure 12] FIG. 11 is a hardware configuration diagram showing hardware of an observation device according to a second embodiment. [Figure 13] 13 is an explanatory diagram showing a comparison result extracted by a score calculation unit 7a and a score Ωn (n=1, . . . , N) of a threshold set Sn based on the extracted comparison result. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] In order to describe the present disclosure in more detail, embodiments of the present disclosure will be described below with reference to the accompanying drawings.

[0010] Embodiment 1 FIG. 1 is a configuration diagram showing an observation device according to the first embodiment. FIG. 2 is a hardware configuration diagram showing the hardware of the observation device according to the first embodiment. The observation device shown in FIG. 1 includes an observation image acquisition unit 1, a threshold set acquisition unit 2, a difference calculation unit 3, a comparison unit 4, a threshold set selection unit 5, and an observation result output unit 6.

[0011] The observed image acquiring unit 1 is realized by, for example, an observed image acquiring circuit 11 shown in FIG. The observation image acquisition unit 1 acquires a first observation image which is an image of an observation target area captured, and a second observation image which is an image of the observation target area captured at a time different from that of the first observation image. The observed image acquisition unit 1 outputs the first observed image and the second observed image to the difference calculation unit 3. The first observation image is, for example, an image taken before an event occurs in the observation target area. The event may be, for example, an earthquake or a flood. The first observation image is preferably an image taken when no abnormality associated with the event occurs. For this reason, the first observation image may be an image taken when the normal situation has been restored after the event has occurred. The second observed image is, for example, an image after an event occurs in the observation target area. The second observed image is not limited to one image, and there may be M observed target images captured at different times. M is an integer equal to or greater than 1.

[0012] The threshold set acquiring unit 2 is realized by, for example, a threshold set acquiring circuit 12 shown in FIG. The threshold set acquisition unit 2 acquires N different threshold sets as threshold sets including thresholds for one or more types of observation components included in both the first observation image and the M observation target images, which are the second observation images, where N is an integer equal to or greater than 2. The threshold set acquisition unit 2 outputs the N threshold sets to each of the comparison unit 4 and the threshold set selection unit 5.

[0013] The difference calculation unit 3 is realized by, for example, a difference calculation circuit 13 shown in FIG. The difference calculation unit 3 acquires the first observed image and the second observed image from the observed image acquisition unit 1, respectively. The difference calculation unit 3 calculates the difference between each observed component included in the first observed image and each observed component included in the second observed image. Specifically, the difference calculation unit 3 calculates the difference between each observation component included in the first observed image and each observation component included in each of the target images. The difference calculation unit 3 outputs the difference between the respective observed components to the comparison unit 4 .

[0014] The comparison unit 4 is realized by, for example, a comparison circuit 14 shown in FIG. The comparison unit 4 acquires N threshold sets from the threshold set acquisition unit 2, and acquires the differences of the respective observed components from the difference calculation unit 3. The comparison unit 4 compares the difference of each observed component with the threshold value for each observed component, which is included in each threshold set. The comparison unit 4 outputs the results of comparing the difference between each observed component and the threshold to the threshold set selection unit 5 and the observation result output unit 6, respectively.

[0015] The threshold set selection unit 5 is realized by, for example, a threshold set selection circuit 15 shown in FIG. The threshold set selection unit 5 includes a score calculation unit 5a and a threshold set selection processing unit 5b. The threshold set selection unit 5 acquires N threshold sets from the threshold set acquisition unit 2, and acquires from the comparison unit 4 the results of comparison between the difference of each observed component and the threshold. The threshold set selection unit 5 selects one of the N threshold sets based on the result of the comparison by the comparison unit 4. The threshold set selection unit 5 outputs the selected threshold set to the observation result output unit 6.

[0016] Based on the result of the comparison by the comparison unit 4, the score calculation unit 5a calculates a score for each of the N threshold value sets as an index for judging the relative merits of the N threshold value sets. Specifically, the score calculation unit 5a acquires local information, which is information indicating whether or not an abnormality has occurred in the observation target area from the outside. The local information is acquired, for example, via a social networking service (SNS) or the Internet of Things (IoT). The score calculation unit 5a compares the local information with the result of the comparison by the comparison unit 4 to determine whether or not the result of the comparison is correct. The score calculation unit 5a calculates the score of each threshold set using the comparison result indicating whether or not the result of the comparison is correct. The score calculation unit 5a outputs the score of each threshold set to the threshold set selection processing unit 5b.

[0017] The threshold set selection processing unit 5b obtains the scores of each threshold set from the score calculation unit 5a. The threshold set selection processing unit 5b compares the N scores calculated by the score calculation unit 5a with each other, and selects one of the N threshold sets based on the score comparison result. The threshold set selection processing unit 5b outputs the selected threshold set to the observation result output unit 6.

[0018] The observation result output unit 6 is realized by, for example, an observation result output circuit 16 shown in FIG. The observation result output unit 6 receives from the comparison unit 4 the comparison result between the difference between each observed component and the threshold value, and receives from the threshold value set selection unit 5 the threshold value set. The observation result output unit 6 extracts, from the results of comparison by the comparison unit 4, the results of comparison in which a threshold included in the threshold set selected by the threshold set selection unit 5 was used. The observation result output unit 6 outputs the extracted comparison result as an observation result, for example, to a display device (not shown).

[0019] In Fig. 1, it is assumed that each of the components of the observation device, that is, an observed image acquisition unit 1, a threshold set acquisition unit 2, a difference calculation unit 3, a comparison unit 4, a threshold set selection unit 5, and an observation result output unit 6, is realized by dedicated hardware as shown in Fig. 2. In other words, it is assumed that the observation device is realized by an observed image acquisition circuit 11, a threshold set acquisition circuit 12, a difference calculation circuit 13, a comparison circuit 14, a threshold set selection circuit 15, and an observation result output circuit 16. Each of the observation image acquisition circuit 11, threshold set acquisition circuit 12, difference calculation circuit 13, comparison circuit 14, threshold set selection circuit 15 and observation result output circuit 16 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a combination of these.

[0020] The components of the observation device are not limited to those realized by dedicated hardware, and the observation device may be realized by software, firmware, or a combination of software and firmware. The software or firmware is stored as a program in the memory of a computer. The computer means hardware that executes the program, and includes, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a central processing unit, a processing unit, an arithmetic unit, a microprocessor, a microcomputer, a processor, or a DSP (Digital Signal Processor).

[0021] FIG. 3 is a hardware configuration diagram of a computer in the case where the observation device is realized by software, firmware, or the like. When the observation device is realized by software, firmware, or the like, a program for causing a computer to execute the respective processing procedures in the observation image acquisition unit 1, the threshold set acquisition unit 2, the difference calculation unit 3, the comparison unit 4, the threshold set selection unit 5, and the observation result output unit 6 is stored in the memory 21. Then, a processor 22 of the computer executes the program stored in the memory 21.

[0022] Also, Fig. 2 shows an example in which each of the components of the observation device is realized by dedicated hardware, and Fig. 3 shows an example in which the observation device is realized by software, firmware, etc. However, this is merely one example, and some of the components in the observation device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0023] Next, the operation of the observation device shown in FIG. 1 will be described. FIG. 4 is a flowchart showing an observation method which is a processing procedure of the observation device. The observation image acquisition unit 1 acquires a first observation image G1 having an observation time t0 as an image obtained by capturing an observation target area (step ST1 in FIG. 4). The observation image acquisition unit 1 also acquires a second observation image G2 with an observation time of t m Observation target image G 2,m (Step ST1 in Fig. 4), where m = 1, . . . , M. FIG. 5 shows a first observation image G1 and an observation target image G 2,m 1 is an explanatory diagram showing each of the above. The observation image acquisition unit 1 acquires a first observation image G1 and an observation target image G 2,m Each of these is output to the difference calculation unit 3.

[0024] The threshold set acquisition unit 2 acquires a first observation image G1 and M observation target images G 2,1 ~G 2,M Each of the observation components OC j The threshold value Th for each of j Let us consider N different threshold sets S 1 ~S N (Step ST2 in FIG. 4). N is an integer equal to or greater than 1, and j=1, . . . , J. J is an integer equal to or greater than 1. One or more observation components OC contained in the first observation image G1, etc. j Examples of the color components include a lightness component, a saturation component, and a hue component. Threshold set Sn (n=1, ,N) is expressed by the following equation (1), which is expressed by J thresholds Th1 n ~Th J n Contains: S 1 =Th1 1 ,Th J 1 S 2 =Th1 2 ,Th J 2 : S N =Th1 N ,Th J N (1) The threshold set acquisition unit 2 acquires N threshold sets S 1 ~S N to the comparison unit 4 and the threshold set selection unit 5, respectively.

[0025] The difference calculation unit 3 receives the first observed image G1 and the M observed target images G 2,1 ~G 2,M Obtain each of the following. The difference calculation unit 3 calculates the observed component OC(1) contained in the first observed image G1 as shown in the following formula (2). j (j=1, ,J) and the observed image G 2,m Observation component OC(2) contained in (m=1, ,M) m,j Difference ΔOC m,j (m = 1, , M; j = 1, , J) is calculated (step ST3 in FIG. 4). OC(1) j and O.C. (2) m,j are the same type of observed components, and the difference between the observed components ΔOC m,j The larger the value, the greater the change in the observed area. ΔOC m,j =|OC(1) j -OC(2) m,j | (2) The difference calculation unit 3 calculates the difference ΔOC m,j is output to the comparison unit 4.

[0026] The comparison unit 4 receives N threshold value sets S 1 ~S N The difference calculation unit 3 obtains the difference ΔOC m,j (m = 1, , M; j = 1, , J). The comparison unit 4 calculates the difference ΔOC m,j and the threshold set S n (n=1, ,N) contains the threshold Th j n (j=1, ,J; n=1, ,N) (step ST4 in FIG. 4). Difference of observed components ΔOC m,j is the threshold Th j n If it is greater than this, the change in the observation area is large, and for example, there is a high possibility that a flood has occurred in the observation area. m,j is the threshold Th j n If it is less than this, the change in the observation target area is small, and for example, there is a low possibility that flooding has occurred in the observation target area. The comparison unit 4 calculates the difference ΔOC m,j and the threshold Th j n Comparison results with R m,j n (m=1,...,M; j=1,...,J; n=1,...,N) is output to the threshold set selection unit 5 and the observation result output unit 6, respectively.

[0027] The score calculation unit 5a of the threshold set selection unit 5 selects N threshold sets S 1 ~S N The comparison unit 4 obtains the observation target image G 2,m The difference ΔOC of the observed components included in (m=1, ,M) m,j Comparison results for R m,j n (m=1,···,M; j=1,···,J; n=1,···,N). The score calculation unit 5a calculates the difference ΔOC between the observed components as shown in the following formula (3). m,jComparison results for R m,j n Based on this, we set N threshold values ​​S 1 ~S N As an index for judging the superiority of n Score Ω for (n=1, ,N) m n (m=1, . . . , M; n=1, . . . , N) is calculated (step ST5 in FIG. 4). The score calculation unit 5a uses a threshold set S n Score Ω m n to the threshold set selection processing unit 5b.

[0028] TIFF0007642181000001.tif27166

[0029] In equation (3), W d u (t d u ) is a weighting factor that changes with time, as shown in FIG. 6. In the example of FIG. 6, d u When the weighting factor W d u (t d u ) reaches its maximum at time t d u When the weighting factor W d u (t d u ) is minimized. Figure 6 shows the weighting factor W d u (t d u ) is an explanatory diagram showing an example of FIG. 7 is an explanatory diagram showing an example of an area for which importance has been set by a user. In FIG. 7, A, B, C, D, and E are areas for which the user has set the importance, and the weighting coefficient W d u (φ d u ,λ d u ,td u ) is the weighting factor for areas A, B, C, D, and E. Weighting factor W d u (φ d u ,λ d u ,t d u ) The subscript d means d=A,B,C,D,E. (φ d u ,λ d u ) indicate the latitude of region d and the longitude of region d, respectively. d u is the user's attention time. W k l is a weighting factor corresponding to local information as shown in FIG. 8, obtained, for example, via SNS or IoT.

[0030] FIG. 8 is an explanatory diagram showing an example of local information indicating whether or not an abnormality has occurred in an observation target area. In the example of Figure 8, a circle indicates that no disaster has occurred at a certain point. For example, W k l = 0.3. × is emergency information that indicates the situation where a disaster is occurring at a certain point. For example, W k l =0.8. In the figure, the numbers 1 to 7 are used to identify certain points. FIG. 9 is an explanatory diagram in which FIG. 7 and FIG. 8 are superimposed. Area A includes points 1 and 2, and K d = 2. Region B includes point 4, and K d = 1. Region C includes point 5, and K d = 1. Region D includes points 6 and 7, and K d =2.

[0031] dt k l is expressed by the following equation (4): 2,m Observation time t mand the time t when local information was obtained k l This shows the time difference between dt k l =t m -t k l (4) W m is the observation time t m is the weighting coefficient. J m,j n is the local information and the comparison result R m,j n The result of the comparison with the local information is shown in Fig. m,j n The comparison result is R m,j n If is correct, then J m,j n is “1”. Local information and comparison result R m,j n The comparison result is R m,j n If is not correct, then J m,j n is "0". For example, the comparison result R corresponding to point 1 m,j n is the threshold Th j n If so, the local information for point 1 is “×”, so J m,j n On the other hand, the comparison result R corresponding to point 1 is m,j n is the threshold Th j n If it is less than 1, the local information of point 1 is “×”, so J m,j n will result in "0". For example, the comparison result R corresponding to point 2 m,j n is the threshold Th j n If so, the local information for point 1 is “〇”, so J m,j n On the other hand, the comparison result R corresponding to point 2 is m,jn is the threshold Th j n If it is less than 1, the local information of point 2 is “〇”, so J m,j n would result in "1". In equation (3), (1-dt k l / W m )J m,j n If it is negative, it is replaced by 0. As described above, the score calculation unit 5a calculates the comparison result R m,j n The result of the match indicates whether the m,j n Using the threshold set S n Score Ω m n Therefore, the threshold set S n Score Ω m n is the threshold set S n The threshold Th included in j n This is an indicator of whether (j = 1, , J; n = 1, , N) is an appropriate value.

[0032] The threshold set selection processing unit 5b selects the threshold set S n Score Ω for (n=1, ,N) m n (m = 1, , M; n = 1, , N). The threshold set selection processing unit 5b selects M observation target images G 2,1 ~G 2,M For each of the N scores Ω m 1 ~Ω m N Compare them to each other. That is, the threshold set selection processing unit 5b selects a threshold set based on, for example, 2,1 If so, the observed image G 2,1 N scores for Ω1 1 ~Ω1 N Compare them to each other. The threshold set selection processing unit 5b is configured to select, for example, 2,M If so, the observed image G 2,M N scores for Ω M 1 ~Ω M N Compare them to each other. The threshold set selection processing unit 5b selects N scores Ω m 1 ~Ω m N Based on the comparison results, we calculate N scores Ω as shown in the following formula (5). m 1 ~Ω m N Among them, the highest score Ω m,MAX Select .

[0033] TIFF0007642181000002.tif15166

[0034] FIG. 10 shows a set of N thresholds S 1 ~S N and the difference between the observed components ΔOC m,j Comparison results for R m,j n and the maximum score Ω m,MAX and the threshold set S m,MAX FIG. A set of N thresholds S 1 ~S N Among them, the highest score is Ω m,MAX A threshold set S corresponding to (m=1, ,M) m,MAX is the observed image G 2,m (m = 1, · · · , M) is likely to be the most suitable set of thresholds. The threshold set selection processing unit 5b selects the maximum score Ω m,MAX The threshold set S corresponding to m,MAX to the observation result output unit 6 (step ST6 in FIG. 4).

[0035] In the observation device shown in FIG. 1, the threshold set selection processing unit 5b selects the maximum score Ω m,MAX The threshold set S corresponding to m,MAXis output to the observation result output unit 6. However, it is sufficient if the threshold set can be optimized, and the threshold set output to the observation result output unit 6 is the maximum score Ω m,MAX The threshold set S corresponding to m,MAX Therefore, the threshold set selection processing unit 5b may select N scores Ω m,1 ~Ω m,N For example, the threshold set corresponding to the second largest score among the above may be output to the observation result output unit 6.

[0036] The observation result output unit 6 outputs the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N) is obtained, and a threshold set S MAX Get the. The observation result output unit 6 outputs the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N) m,MAX The J thresholds Th1 included in n ~Th J n The comparison result in which the comparison result is used is extracted (step ST7 in FIG. 4). For example, the threshold set S m,MAX is the threshold set S 3 If so, the observation result output unit 6 outputs the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N) 3 The J thresholds Th1 included in 3 ~Th J 3 Comparison results using R m,j 3 Extract. For example, the threshold set S m,MAX is the threshold set S 4 If so, the observation result output unit 6 outputs the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N)4 The J thresholds Th1 included in 4 ~Th J 4 Comparison results using R m,j 4 Extract. The observation result output unit 6 outputs the extracted comparison result as an observation result, for example, to a display device (not shown).

[0037] In the above-described first embodiment, the observation device is configured to include a threshold set acquisition unit 2 that acquires N (N is an integer of 2 or more) different threshold sets as threshold sets including thresholds for one or more types of observation components included in both a first observation image, which is an image of the observation target area captured, and a second observation image, which is an image of the observation target area captured at a time different from that of the first observation image, and a difference calculation unit 3 that calculates a difference between each observation component included in the first observation image and each observation component included in the second observation image. The observation device also includes a comparison unit 4 that compares the difference between each observation component calculated by the difference calculation unit 3 with the threshold for each observation component included in each threshold set acquired by the threshold set acquisition unit 2, and a threshold set selection unit 5 that selects one of the N threshold sets acquired by the threshold set acquisition unit 2 based on the result of the comparison by the comparison unit 4. Therefore, the observation device can improve the detection accuracy of changes in the observation target area compared to the observation device disclosed in Patent Document 1.

[0038] Embodiment 2 In the second embodiment, an observation device in which the threshold set selection unit 7 includes a score calculation unit 7a and a threshold set selection processing unit 7b will be described.

[0039] Fig. 11 is a configuration diagram showing an observation device according to embodiment 2. In Fig. 11, the same reference numerals as in Fig. 1 indicate the same or corresponding parts, and detailed explanations will be omitted. Fig. 12 is a hardware configuration diagram showing the hardware of an observation device according to embodiment 2. In Fig. 12, the same reference numerals as in Fig. 2 indicate the same or corresponding parts, and therefore detailed description thereof will be omitted. The observation device shown in FIG. 11 includes an observation image acquisition unit 1, a threshold set acquisition unit 2, a difference calculation unit 3, a comparison unit 4, a threshold set selection unit 7, and an observation result output unit 6.

[0040] The threshold set selector 7 is realized by, for example, a threshold set selector circuit 17 shown in FIG. The threshold set selection unit 7 includes a score calculation unit 7a and a threshold set selection processing unit 7b. The threshold set selection unit 7 acquires N threshold sets from the threshold set acquisition unit 2, and acquires from the comparison unit 4 the results of comparison between the difference of each observed component and the threshold. The threshold set selection unit 7 selects one of the N threshold sets based on the result of the comparison by the comparison unit 4. The threshold set selection unit 7 outputs the selected threshold set to the observation result output unit 6.

[0041] The score calculation unit 7a receives the comparison result R m,j n (m=1,···,M; j=1,···,J; n=1,···,N). The score calculation unit 7a calculates the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N) for each threshold set S n Extract the comparison results where (n = 1, , N) was used. The score calculation unit 7a calculates the score for each threshold set S based on the extracted comparison result. n Score Ω n Calculate (n = 1, , N). The score calculation unit 7a uses a threshold set S n Score Ω n (n=1, . . . , N) is output to threshold set selection processing unit 7b.

[0042] The threshold set selection processing unit 7b selects the threshold set S n Score Ω n Obtain (n=1, ,N). The threshold set selection processing unit 7b selects N scores Ω 1 ~Ω N are compared with each other and the score Ω 1 ~Ω N Based on the comparison results, N threshold sets S 1 ~S N Select one of the threshold sets from among The threshold set selection processing unit 7 b outputs the selected threshold set to the observation result output unit 6 .

[0043] In Fig. 11, it is assumed that each of the components of the observation device, that is, the observed image acquisition unit 1, the threshold set acquisition unit 2, the difference calculation unit 3, the comparison unit 4, the threshold set selection unit 7, and the observation result output unit 6, is realized by dedicated hardware as shown in Fig. 12. In other words, it is assumed that the observation device is realized by an observed image acquisition circuit 11, a threshold set acquisition circuit 12, a difference calculation circuit 13, a comparison circuit 14, a threshold set selection circuit 17, and an observation result output circuit 16. Each of the observation image acquisition circuit 11, the threshold set acquisition circuit 12, the difference calculation circuit 13, the comparison circuit 14, the threshold set selection circuit 17 and the observation result output circuit 16 corresponds to, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination of these.

[0044] The components of the observation device are not limited to those realized by dedicated hardware, and the observation device may be realized by software, firmware, or a combination of software and firmware. When the observation device is realized by software, firmware, or the like, a program for causing a computer to execute the respective processing procedures in the observation image acquisition unit 1, the threshold set acquisition unit 2, the difference calculation unit 3, the comparison unit 4, the threshold set selection unit 7, and the observation result output unit 6 is stored in a memory 21 shown in Fig. 3. Then, a processor 22 shown in Fig. 3 executes the program stored in the memory 21.

[0045] Also, Fig. 12 shows an example in which each of the components of the observation device is realized by dedicated hardware, while Fig. 3 shows an example in which the observation device is realized by software, firmware, etc. However, this is merely one example, and some of the components in the observation device may be realized by dedicated hardware, and the remaining components may be realized by software, firmware, etc.

[0046] Next, the operation of the observation device shown in Fig. 11 will be described. However, other than the score calculation unit 7a and the threshold set selection processing unit 7b, the observation device is the same as that shown in Fig. 1. Therefore, here, only the operation of the score calculation unit 7a and the threshold set selection processing unit 7b will be described.

[0047] The score calculation unit 7a receives the comparison result R m,j n (m=1,···,M; j=1,···,J; n=1,···,N). The score calculation unit 7a calculates the comparison result R m,j n (m=1, ,M; j=1, ,J; n=1, ,N) for each threshold set S n Extract the comparison results where (n = 1, , N) was used. The comparison results extracted by the score calculation unit 7a are, as shown in FIG. 2,1 ~G 2,M J observation components OC(2) contained in each of m,j Among the comparison results for (m=1, ,M;j=1, ,J), each threshold set S n(n=1, ,N) was used for the comparison results. In Fig. 13, the threshold set S 1 Comparison results using R 1,j 1 ~R M,j 1 and the threshold set S 2 Comparison results using R 1,j 2 ~R M,j 2 and the threshold set S N Comparison results using R 1,j N ~R M,j N An example is shown in which the following is extracted: FIG. 13 shows the comparison result extracted by the score calculation unit 7a and the threshold value set S based on the extracted comparison result. n Score Ω n FIG. 1 is an explanatory diagram showing (n=1, . . . , N).

[0048] The score calculation unit 7a calculates each threshold set S based on the extracted comparison result as shown in the following formula (6). n Score Ω n Calculate (n = 1, , N). The score calculation unit 7a uses a threshold set S n Score Ω n (n=1, . . . , N) is output to threshold set selection processing unit 7b.

[0049] TIFF0007642181000003.tif25166

[0050] The threshold set selection processing unit 7b selects the threshold set S n Score Ω n Obtain (n=1, ,N). The threshold set selection processing unit 7b selects N scores Ω 1 ~Ω N are compared with each other, and N scores Ω are calculated as shown in the following formula (7). 1 ~Ω N Among them, the highest score is Ω MAXSelect . The threshold set selection processing unit 7b selects N threshold sets S 1 ~S N Among them, the highest score Ω MAX The threshold set S corresponding to MAX is output to the observation result output unit 6.

[0051] TIFF0007642181000004.tif11166

[0052] In the observation device shown in FIG. 11, the threshold set selection processing unit 7b selects the maximum score Ω MAX The threshold set S corresponding to MAX is output to the observation result output unit 6. However, if the threshold set can be optimized, the threshold set output to the observation result output unit 6 should be the maximum score Ω MAX The threshold set S corresponding to MAX Therefore, the threshold set selection processing unit 7b may select N scores Ω 1 ~Ω N For example, the threshold set corresponding to the second largest score among the above may be output to the observation result output unit 6.

[0053] In the above-described second embodiment, the observation device is configured to include a threshold set selection unit 7 that extracts comparison results in which each threshold set acquired by the threshold set acquisition unit 2 is used from the comparison results by the comparison unit 4 and calculates a score for each threshold set based on the extracted comparison results, and a threshold set selection processing unit 7b that compares the multiple scores calculated by the score calculation unit 7a with each other and selects one of the N threshold sets acquired by the threshold set acquisition unit 2 based on the score comparison results. Therefore, the observation device can improve the detection accuracy of changes in the observation target area compared to the observation device disclosed in Patent Document 1.

[0054] In addition, the present disclosure allows free combination of the respective embodiments, modification of any of the components of each embodiment, or omission of any of the components of each embodiment. [Industrial Applicability]

[0055] The present disclosure is suitable for an observation device and an observation method. [Explanation of symbols]

[0056] 1 Observation image acquisition unit, 2 Threshold set acquisition unit, 3 Difference calculation unit, 4 Comparison unit, 5 Threshold set selection unit, 5a Score calculation unit, 5b Threshold set selection processing unit, 6 Observation result output unit, 7 Threshold set selection unit, 7a Score calculation unit, 7b Threshold set selection processing unit, 11 Observation image acquisition circuit, 12 Threshold set acquisition circuit, 13 Difference calculation circuit, 14 Comparison circuit, 15 Threshold set selection circuit, 16 Observation result output circuit, 17 Threshold set selection circuit, 21 Memory, 22 Processor.

Claims

1. a threshold set acquisition unit that acquires N (N is an integer equal to or greater than 2) different threshold sets as threshold sets including thresholds for one or more types of observation components included in both a first observation image, which is an image of an observation target area captured, and a second observation image, which is an image of the observation target area captured at a time different from that of the first observation image; a difference calculation unit that calculates a difference between each observed component included in the first observed image and each observed component included in the second observed image; a comparison unit that compares the differences of the respective observation components calculated by the difference calculation unit with thresholds for the respective observation components included in the respective threshold sets acquired by the threshold set acquisition unit; a threshold set selection unit that selects one of the N threshold sets acquired by the threshold set acquisition unit based on a result of the comparison by the comparison unit; An observation device equipped with

2. The observation device according to claim 1, further comprising an observation result output unit that extracts comparison results in which thresholds included in the threshold set selected by the threshold set selection unit are used from the comparison results by the comparison unit, and outputs the extracted comparison results as observation results.

3. The threshold set selection unit: a score calculation unit that calculates a score for each of the N threshold sets as an index for determining superiority or inferiority of the N threshold sets acquired by the threshold set acquisition unit based on a result of the comparison by the comparison unit; The observation device according to claim 1, further comprising a threshold set selection processing unit that compares the N scores calculated by the score calculation unit with each other and selects a threshold set from the N threshold sets acquired by the threshold set acquisition unit based on a result of the score comparison.

4. The score calculation unit is The observation device described in claim 3, characterized in that it acquires local information from the outside, which is information indicating whether or not an abnormality has occurred in the observation target area, compares the local information with the results of the comparison by the comparison unit, and calculates a score for each threshold set using the results of the comparison.

5. As the second observation image, there are M (M is an integer equal to or greater than 1) observation target images having different photographing times, The difference calculation unit Calculating a difference between each observation component included in the first observation image and each observation component included in each observation target image; The comparison unit is The observation device according to claim 1, characterized in that the difference of each observation component calculated by the difference calculation unit is compared with a threshold value for each observation component included in the N threshold sets acquired by the threshold set acquisition unit.

6. The threshold set selection unit: a score calculation unit that calculates a score for each of the N threshold sets as an index for determining superiority or inferiority of the N threshold sets acquired by the threshold set acquisition unit based on a result of the comparison by the comparison unit; The observation device according to claim 5, further comprising a threshold set selection processing unit that compares the N scores calculated by the score calculation unit with each other and selects a threshold set from the N threshold sets acquired by the threshold set acquisition unit based on a result of the score comparison.

7. The threshold set selection unit: a score calculation unit that extracts comparison results in which each of the threshold sets acquired by the threshold set acquisition unit is used from the comparison results by the comparison unit, and calculates a score for each of the threshold sets based on the extracted comparison results; The observation device according to claim 5, further comprising a threshold set selection processing unit that compares the multiple scores calculated by the score calculation unit with each other and selects a threshold set from among the N threshold sets acquired by the threshold set acquisition unit based on a result of the score comparison.

8. a threshold set acquisition unit acquires N (N is an integer equal to or greater than 2) threshold sets different from each other as threshold sets including thresholds for one or more types of observation components included in both a first observation image which is an image of an observation target area captured, and a second observation image which is an image of the observation target area captured at a time different from that of the first observation image; a difference calculation unit calculates a difference between each observed component included in the first observed image and each observed component included in the second observed image; a comparison unit that compares the difference between each of the observation components calculated by the difference calculation unit with a threshold value for each of the observation components included in each of the threshold sets acquired by the threshold set acquisition unit; A threshold set selection unit selects one of the N threshold sets acquired by the threshold set acquisition unit based on a result of the comparison by the comparison unit. Observation method.

Images