JP7105896B2 - Window detection device, air conditioning control device, air conditioning system, window detection method, and program - Google Patents

Description

The present invention relates to a window detection device, an air conditioning control device, an air conditioning system , a window detection method , and a program for detecting a window area from a target area.

2. Description of the Related Art Air conditioners are known that can determine conditions such as the environment in a room and the position of an occupant, and control the direction of airflow, output, etc. according to the determined conditions. In such an air conditioner, it is important to know the position, size, temperature, etc. of windows through which cold air and warm air can escape more easily than other parts.

Here, Patent Documents 1 to 3 are known as techniques for detecting windows from inside a room. Japanese Patent Application Laid-Open No. 2002-200001 discloses a technique for determining a window region from the ambient temperature difference in thermal image data. Patent Literatures 2 and 3 disclose techniques for detecting a window region from a difference in brightness indicated by near-infrared image data.

JP 2013-64598 A JP 2016-200282 A JP 2017-44373 A

Unlike ordinary houses, office spaces are equipped with a large number of heat-generating soup bowls and equipment. In addition, there are many residents in the office space, and the environment of the office space often changes greatly depending on the time of day. Therefore, when detecting a window region in an area such as an office space using the methods described in Patent Documents 1 to 3, there is a possibility that the window region cannot be detected correctly.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a window detection device, an air-conditioning control device, an air-conditioning system, a window detection method, and a program, which are capable of detecting a window area with higher accuracy than in the past. and

In order to achieve the above object, the window detection device according to the present invention includes:
Acquisition means for acquiring thermal image data of a target region over a predetermined period;
window detection means for detecting a window region within the target region from a plurality of the thermal image data acquired by the acquisition means and having different acquisition times;
with
The window detection means is
window discrimination execution means for repeatedly executing a window discrimination process for discriminating whether or not each divided region obtained by dividing the search region included in the target region is a window region until a predetermined end condition is satisfied;
Search area setting means for setting the search area to be targeted in the next window discrimination process based on the discrimination result of the window discrimination process,
The window determination executing means includes, as the window determination process, a first window determination process and a second window determination process which is executed after the first window determination process and determines the window region by a method different from the first window determination process. and are repeatedly executed.

According to the present invention, it is possible to detect a window region with higher accuracy than conventionally.

A diagram showing the configuration of an air conditioning system according to an embodiment of the present invention. 1 is a block diagram showing the configuration of an air conditioning control device according to an embodiment of the present invention; FIG. FIG. 2 is a diagram showing information stored in the secondary storage device of the air conditioning control device according to the embodiment of the present invention; 1 is a functional configuration diagram of an air conditioning control device according to an embodiment of the present invention; Flowchart of window detection processing according to the embodiment of the present invention Diagram showing an example of setting a search area (Part 1) Diagram showing an example of setting a search area (Part 2) Flowchart of first window determination processing according to the embodiment of the present invention Diagram for explaining the first window determination process (Part 1) Diagram for explaining the first window determination process (Part 2) Flowchart of second window determination processing according to the embodiment of the present invention Diagram for explaining the second window determination process (Part 1) Diagram for explaining the second window determination process (Part 2) Flowchart of air conditioning control processing according to the embodiment of the present invention

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The same reference numerals are given to the same or corresponding parts in the drawings.

FIG. 1 is a diagram showing the overall configuration of an air conditioning system 100 according to an embodiment of the invention. The air conditioning system 100 includes an air conditioner 10 for air conditioning an office space 200 , an infrared camera 20 , and an air conditioning control device 30 . The air conditioner 10 and the infrared camera 20 are communicably connected to the air conditioning control device 30 via the network N1. The network N1 is, for example, a LAN (Local Area Network), the Internet, or the like.

The air conditioner 10 is installed in the office space 200 and air-conditions the office space 200 by performing cooling operation, heating operation, dehumidification operation, etc. based on instructions from the air conditioning control device 30 .

The infrared camera 20 is installed in the vicinity of the air conditioner 10, and captures thermal image data indicating the temperature distribution of the imageable area at predetermined time intervals (for example, every minute) and transmits the thermal image data to the air conditioning control device 30. . In this embodiment, as shown in FIG. 1, the infrared camera 20 is installed at a position facing a wall 202 having a window 201, and the imageable area includes the entire wall 202. As shown in FIG. In the following description, the imageable area of the infrared camera 20 is also referred to as a target area. A plurality of infrared cameras 20 installed at different positions may be provided so as to cover the entire office space 200 . The infrared camera 20 is an example of the imaging device of the present invention.

The air conditioning control device 30 is communicably connected to the air conditioner 10 and the infrared camera 20 via the network N1. The air conditioning control device 30 detects the window area based on the thermal image data acquired from the infrared camera 20 and acquired at different times, and controls the operation of the air conditioner 10 based on the detection result. The air conditioning control device 30 is an example of the window detection device of the present invention.

Next, the configuration of the air conditioning control device 30 will be described. The air conditioning control device 30 includes a communication interface 31, a processor 32, a ROM 33, a RAM 34, and a secondary storage device 35, as shown in FIG. These components are interconnected via bus 36 .

The communication interface 31 has an interface for communicating with the air conditioner 10 and the infrared camera 20 via the network N1.

A processor 32 comprehensively controls the air conditioning control device 30 . A ROM (Read Only Memory) 33 stores a plurality of pieces of firmware and data used when these pieces of firmware are executed. A RAM (Random Access Memory) 34 is used as a work area for the processor 32 .

The secondary storage device 35 includes a readable/writable non-volatile semiconductor memory such as EEPROM, flash memory, or HDD. As shown in FIG. 3, the secondary storage device 35 stores a window detection program 351, an air conditioning control program 352, an image DB 353, and a determination result DB 354.

The air-conditioning control program 352 is a program describing a process of detecting a window area from a plurality of thermal image data acquired from the infrared camera 20 at different acquisition times. The air conditioning control program 352 is a program in which processing for controlling the air conditioner 10 based on the detection result of the window area is described.

The image DB 353 cumulatively stores the thermal image data of the target area acquired from the infrared camera 20 in association with the date and time of acquisition.

The determination result DB 354 stores determination result information indicating the result of window determination detected by a window detection process, which will be described later. The determination result information is information indicating, for each small area (for example, each pixel) in the target area, the accuracy of whether or not the small area is a window area (hereinafter referred to as window discrimination accuracy). In this embodiment, the window discrimination accuracy is expressed in 10 stages from 1 to 10. FIG. A window discrimination accuracy of "1" indicates that the small area is unlikely to be a window area. A window discrimination accuracy of "10" indicates that the small area is highly likely to be a window area. A window discrimination accuracy of "5" indicates that it is unclear whether or not the small area is a window area. Note that the window discrimination accuracy may be calculated in finer steps, or the window discrimination accuracy may be expressed by the probability of whether or not it is a window region.

FIG. 4 is a block diagram showing the functional configuration of the air conditioning control device 30. As shown in FIG. The air conditioning control device 30 functionally includes an acquisition unit 301 , a window detection unit 302 , an open/close determination unit 303 , and an air conditioning control unit 304 . The acquisition unit 301 and the window detection unit 302 are implemented by the processor 32 executing a window detection program 351 stored in the secondary storage device 35 . The open/close determination unit 303 and the air conditioning control unit 304 are implemented by the processor 32 executing an air conditioning control program 352 stored in the secondary storage device 35 .

The acquisition unit 301 acquires the thermal image data captured by the infrared camera 20, and accumulates and stores it in the image DB 353 in association with the information indicating the acquisition time. Acquisition unit 301 is an example of acquisition means of the present invention. The operation of the acquisition unit 301 is an example of the acquisition step of the present invention.

The window detection unit 302 detects a window region within the target region from a plurality of thermal image data with different acquisition times stored in the image DB 353 and updates the determination result information in the determination result DB 354 . Window detection unit 302 is an example of the window detection means of the present invention. The operation of the window detection unit 302 is an example of the window detection step of the present invention. The window detection unit 302 includes a window determination execution unit 3021 and a search area setting unit 3022 as a more detailed configuration.

The window determination execution unit 3021 repeats the window determination process for determining whether or not the divided area is a window area for each divided area obtained by dividing the search area included in the target area until a predetermined end condition is satisfied. Run. The window discrimination executing section 3021 is an example of the window discrimination executing means of the present invention.

In this embodiment, the window discrimination execution unit 3021 repeatedly executes the first window discrimination process and the second window discrimination process, which are two kinds of window discrimination processes with different algorithms, until the termination condition is satisfied. The first window determination process is a process of detecting a window region based on temperature distribution conditions indicated by a plurality of thermal image data acquired by the acquisition means. The second window determination process is a process of detecting a window region based on temporal temperature changes indicated by a plurality of thermal image data acquired by the acquisition means.

Here, the first window discrimination process discriminates a window region based on a temperature distribution state that can be obtained by summing temperatures indicated by a plurality of thermal image data for each constant temperature range regardless of time change. Therefore, the second window determination process for detecting the window region based on the temperature change over time has a larger processing load than the first window determination process. Details of the processing performed by the window determination execution unit 3021 will be described later.

The search area setting unit 3022 performs processing for setting search areas and divided areas to be targeted in the next window discrimination process based on the execution result of the window discrimination process. Details of the processing performed by the search area setting unit 3022 will be described later. The search area setting section 3022 is an example of the search area setting means of the present invention.

The open/close determination unit 303 determines the open/closed state of the detected window region. The details of the processing performed by the open/close determination unit 303 will be described later. The open/close determination unit 303 is an example of the open/close determination unit of the present invention.

The air conditioning control unit 304 controls the air conditioner 10 based on the determination result information stored in the determination result DB 354, the latest thermal image data stored in the image DB 353, the determination result by the open/close determination unit 303, and the like. The air conditioning control unit 304 is an example of air conditioning control means of the present invention. Details of the processing performed by the air conditioning control unit 304 will be described later.

Next, the operation of window detection processing executed by the air conditioning control device 30 will be described. For example, when the air conditioning control device 30 is connected to the air conditioner 10 and the infrared sensor 20 via the network N1 for the first time, or when an instruction signal to start processing is received from an administrator via a remote controller (not shown), the air conditioning The control device 30 executes window detection processing shown in FIG.

First, the acquiring unit 301 periodically acquires thermal image data captured by the infrared camera 20 for a predetermined aggregation period, and stores the thermal image data in the image DB 353 (step S101). It should be noted that this totaling period is at least one day or more, preferably several days. The aggregation period is an example of a predetermined period of the present invention.

After obtaining the thermal image data for the aggregation period, the search area setting unit 3022 sets the search area to be determined in the first window determination process described later based on the determination result information stored in the determination result DB 354. (Step S102). Specifically, based on the determination result information, the search area setting unit 3022 sets the search area so as to include an area where it is uncertain whether or not it is a window area. It should be noted that in the first process, the determination result information is not stored in the determination result DB 354, so the search area setting unit 3022 sets the entire target area as the search area.

Here, the processing of step S102 will be described with an example. FIG. 6A is a diagram showing the window discrimination accuracy of the target region indicated by the current determination result information. In this case, the search area setting unit 3022 sets the hatched area shown in FIG. 6B, which is an area other than the window discrimination accuracy of "1" and "10", as the search area. Note that the search area setting unit 3022 may set the search area including not only the area where it is uncertain whether it is a window area, but also the area around it.

Returning to FIG. 5, the search area setting unit 3022 subsequently divides the search area set in step S102 into divided areas of the same size, the number of divisions being N (step S103). Note that the process of step S103 may be repeatedly executed, but the number of divisions N is the same each time.

Subsequently, the window discrimination execution unit 3021 executes first window discrimination processing for discriminating a window region from the target region based on a plurality of thermal image data acquired during the tallying period stored in the image DB 353 (step S104). ). Details of the first window determination process will be described with reference to the flowchart of FIG.

First, the window determination execution unit 3021 selects one unselected region from among the divided regions divided in step S103 (step S1041). For ease of explanation, the divided area selected in step S1041 is also referred to as a selected area in the following explanation.

Subsequently, the window discrimination executing unit 3021 calculates the average temperature of the selected area for each of the plurality of thermal image data acquired during the aggregation period stored in the image DB 353 (step S1042).

Subsequently, the window discrimination execution unit 3021 aggregates the average temperature of the selected region calculated for each thermal image data for each predetermined temperature range, and obtains information indicating the temperature distribution (step S1043). The information indicating the temperature distribution state is, for example, average, variance, number of peaks, and the like.

Then, the window discrimination execution unit 3021 calculates the window discrimination accuracy of the selected region based on the information indicating the temperature distribution (step S1044).

Here, an example of calculating the window discrimination accuracy in step S1044 will be described. For example, the window discrimination execution unit 3021 discriminates whether or not the average, variance, and number of peaks, which are information indicating the temperature distribution state, match conditions 1 to 3 shown below. Note that the number of peaks in Condition 1 may be calculated using a known peak analysis technique. The average room temperature of condition 3 may be calculated from the measured value of a temperature sensor (not shown) provided in the air conditioner 10, the measured value of a room temperature sensor (not shown) separately provided in the office space 200, and the like.
(Condition 1) The number of peaks is 2 or more.
(Condition 2) Variance is greater than the threshold.
(Condition 3) The difference between the average and the average room temperature in the aggregation period is greater than the threshold

Then, the window discrimination execution unit 3021 may obtain the window discrimination accuracy of the selected region based on the number of times that the conditions 1 to 3 are met. For example, the window discrimination execution unit 3021 sets the window discrimination accuracy to "1" if not even one match, the window discrimination accuracy to "2" if only one matches, and the window discrimination accuracy to "2" if two matches. The window discrimination accuracy may be calculated such as "5" and "10" when three match.

For example, consider a case where the temperature distribution of the selected area is as shown in FIG. 8A. In this case, according to the calculation example of the window discrimination accuracy described above, since the number of peaks is 2, condition 1 is met, and since the variance is large and above the threshold, condition 2 is also met, so condition 3 is also met. If so, the possibility that the selected area is the window area is high, and a relatively high window discrimination accuracy is calculated.

On the other hand, when the temperature distribution of the selected region is as shown in FIG. is a window region, and a relatively low window discrimination accuracy is calculated.

Returning to FIG. 7, the window discrimination executing unit 3021 updates the value of the window discrimination accuracy of the selected region indicated by the discrimination result information stored in the discrimination result DB 354 based on the window discrimination accuracy calculated in step S1044 (step S1045).

Specifically, the window discrimination execution unit 3021 may update the value of the window discrimination accuracy of the selected region to the value calculated by the following formula (1). α in equation (1) is a count ranging from 0 to 1.
Formula (1): (window discrimination accuracy of selected area after update)=(window discrimination accuracy of selected area indicated by current determination result information)*(1−α)+(window discrimination of selected area calculated in step S1044) Accuracy) *α

Subsequently, the window discrimination execution unit 3021 discriminates whether or not all divided regions have been selected in step S1041 (step S1046). If there is an unselected divided area (step S1046; No), the window discrimination execution unit 3021 selects the unselected divided area and calculates the window discrimination accuracy of the selected divided area (that is, the selected area). The process of updating the determination result information (steps S1041 to S1045) is repeated. If there is no unselected divided area (step S1046; Yes), the first window determination process ends.

Returning to FIG. 5, when the first window determination process ends, the window detection unit 302 determines whether or not a predetermined end condition is satisfied (step S105). Specifically, the window detection unit 302 detects that the values of the window discrimination accuracy for each small region indicated by the determination result information stored in the determination result DB 354 are only “1” and “10” (that is, the window region If there is no region in which it is uncertain whether or not the condition is satisfied, it may be determined that the termination condition is satisfied. Alternatively, the window detection unit 302 detects when the number of repetitions of the first window discrimination processing reaches a predetermined number even when there are regions with window discrimination accuracies other than "1" and "10" in the target region. may determine that the termination condition is satisfied.

If the end condition is satisfied (step S105; Yes), the window detection process ends. On the other hand, if the end condition is not satisfied (step S105; No), the search area setting unit 3022 performs the second window determination process, which will be described later, based on the determination result information stored in the determination result DB 354, as in step S102. , a search area to be determined is set (step S106). Then, the search area setting unit 3022 divides the set search area into N divided areas of the same size, as in step S103 (step S107).

Subsequently, the window discrimination executing unit 3021 discriminates the window region from the target region by a method different from the first window discrimination process, based on a plurality of thermal image data acquired during the aggregation period stored in the image DB 353. 2-window discrimination processing is executed (step S108). Details of the second window determination process will be described with reference to the flowchart of FIG.

First, the window determination executing unit 3021 selects one unselected region from among the divided regions divided in step S107 (step S1081). For ease of explanation, the divided area selected in step S1081 is also referred to as a selected area in the following explanation. Then, the window discrimination executing unit 3021 calculates the average temperature of the selected area for each of the plurality of thermal image data acquired during the aggregation period stored in the image DB 353 (step S1082).

Subsequently, the window determination execution unit 3021 applies the temporal change in the average temperature of the selected region calculated for each thermal image data to an approximated curve using a known method such as the least squares method or polynomial approximation (step S1083). ).

Then, the window discrimination execution unit 3021 calculates a correlation coefficient indicating the degree of correlation between the fitted approximated curve and the actual average temperature using a known method, and calculates the window discrimination accuracy of the selected region from the calculated correlation coefficient. is obtained (step S1084). Here, it is empirically known that the window region changes temperature more smoothly than other regions. Therefore, the window discrimination execution unit 3021 may obtain the window discrimination accuracy according to a predetermined criterion such that the larger the correlation coefficient, the higher the value.

For example, the change over time of the average temperature in the selected region indicated by the solid line in the graph shown in FIG. 10A deviates greatly from the approximate curve obtained from this change over time indicated by the dotted line, and the correlation coefficient is low. Therefore, the window discrimination accuracy of the selected region in this case is calculated to be low.

On the other hand, the time change of the average temperature of the selected region indicated by the solid line in the graph shown in FIG. 10B does not deviate much from the approximate curve obtained from this time change indicated by the dotted line, and the correlation coefficient is large. Therefore, the window discrimination accuracy of the selected region in this case is calculated to be high.

Returning to FIG. 9, subsequently, the window discrimination executing unit 3021 updates the value of the window discrimination accuracy of the selected region indicated by the discrimination result information stored in the discrimination result DB 354 based on the window discrimination accuracy calculated in step S1084. (step S1085).

Subsequently, the window discrimination execution unit 3021 discriminates whether or not all divided regions have been selected in step S1081 (step S1086). If there is an unselected divided area (step S1086; No), the window discrimination execution unit 3021 selects the unselected divided area and calculates the window discrimination accuracy of the selected divided area (that is, the selected area). The process of updating the determination result information (steps S1081 to S1085) is repeated. If there is no unselected divided area (step S1086; Yes), the second window determination process ends.

Returning to FIG. 5, when the second window determination process ends, the window detection unit 302 determines whether or not a predetermined termination condition is satisfied (step S109). Note that this termination condition may be the same as the termination condition of step S105, or may be a different condition.

If the termination condition is not satisfied (step S109; No), the process returns to step S101, and the window detection unit 302 performs processing for determining a window region based on the thermal image data acquired during the aggregation period (steps S101 to S108). repeat. On the other hand, if the end condition is satisfied (step S109; Yes), the window detection process ends.

By executing the window detection process in this way, the window region can be accurately detected from a plurality of thermal image data for the aggregation period in which the target region is imaged by the infrared camera 20 .

Next, the operation of the air conditioning control process executed by the air conditioning control device 30 will be described. Note that the window detection process has already been executed before the air conditioning control process is started. For example, the air-conditioning control device 30 executes the air-conditioning control process shown in FIG. 11 every predetermined time (for example, 5 minutes).

First, the air conditioning control unit 304 determines the open/closed state of the window region indicated by the determination result information stored in the determination result DB 354 (step S201). For example, the air-conditioning control unit 304 acquires the latest thermal image data from the infrared camera 20, checks whether there is a portion with a different temperature protruding into the window area, and if there is such a portion, the window is closed. is open, and if there is no such portion, it is determined that the window is closed. Alternatively, if the brightness in the target area can be measured by a brightness sensor (not shown), the air conditioning control unit 304 determines that the window is closed if there is a portion with brightness higher than the threshold value in the window area, and determines that the window is closed. If there is no such portion, it may be determined that the window is open. Any method may be used to determine the open/closed state of the window region, and another method may be used to determine the open/closed state.

Subsequently, the air conditioning control unit 304 controls the air conditioner 10 based on the position of the window area, the open/close state, etc. (step S202). For example, when the current operation mode of the air conditioner 10 is the energy saving mode and the window is open, the air conditioning control unit 304 transmits a stop command to the air conditioner 10 via the network N1, and the air conditioner 10 stop the operation of Further, when the window is closed, the air conditioning control unit 304 obtains the difference between the average temperature of the window area calculated from the latest thermal image data and the set temperature of the air conditioner 10, and if the difference is larger than the threshold, the air volume is transmitted to the air conditioner 10 via the network N1. The air-conditioning control process ends here.

As described above, according to the air-conditioning control device 30 according to the present embodiment, the window region within the target region is detected from a plurality of thermal image data acquired over a predetermined period. Therefore, even when the detection of the window area is performed for an area such as the office space 200 in which a large number of heat-generating soup bowls, equipment, etc. are installed and the environment changes greatly depending on the time of day, a plurality of windows can be detected. It is possible to correctly determine the window area from the thermal image data and detect the window area with higher accuracy than in the past.

Further, the air-conditioning control device 30 according to the present embodiment executes a first window determination process of detecting a window region based on the temperature distribution conditions indicated by multiple pieces of thermal image data. The air-conditioning control device 30 also executes a second window determination process for detecting a window region based on temperature changes over time indicated by a plurality of thermal image data. The air-conditioning control device 30 detects the window area by repeatedly executing the first window determination process and the second window determination process with different algorithms until the end condition is satisfied. Then, when repeating the window discrimination process in this manner, the search area for the next window discrimination process is set based on the execution result of the previous window discrimination process. Therefore, in the present embodiment, since the window determination process is repeatedly executed with emphasis on the area where it is uncertain whether it is a window, the window area can be accurately determined while suppressing the processing load. becomes possible.

In addition, in the present embodiment, the second window determination process, which has a larger processing load than the first window determination process, is performed after the first window determination process is performed, so it is possible to further reduce the processing load. .

(Modification)
It should be noted that the present invention is not limited to the above-described embodiments, and various modifications are of course possible without departing from the gist of the present invention.

In the above embodiment, as the window detection process, the first window discrimination process and the second window discrimination process are repeatedly executed until the end condition is satisfied. may be executed repeatedly. Also, if it is desired to roughly determine the window region without spending time on the processing, the first window determination processing and/or the second window determination processing are executed only once without such repetition, and the window detection processing is performed. may be terminated.

In the above embodiment, the first window determination process is a process of detecting a window region based on the temperature distribution conditions indicated by a plurality of thermal image data. Further, it has been described that the second window determination process is a process of detecting a window region based on temporal temperature changes indicated by a plurality of thermal image data. However, as long as a plurality of thermal image data with different acquisition times are used, various algorithms can be adopted for the first window discrimination process and the second window discrimination process. For example, as the first window discrimination process and the second window discrimination process, the average temperature is calculated for each divided area from the thermal image data acquired at different times, and it is compared with a threshold to determine whether it is a window area or not. processing may be performed.

Further, in the above embodiment, the thermal image data is stored in the image DB 353, and the window area is detected based on a plurality of thermal image data stored in the window detection process. However, when the thermal image data is acquired, the information used for window discrimination (for example, the average temperature of each divided area) is calculated in the first window discrimination process and the second window discrimination process, and only the calculated information is stored. You may make it By doing so, the capacity of the secondary storage device 35 can be saved.

Further, in the above embodiment, by applying the window detection program 351 and the air conditioning control program 352 executed by the processor 32 of the air conditioning control device 30 to an existing computer, the computer functions as the air conditioning control device 30 according to the present invention. It is also possible to let

Any method can be used to distribute such a program. may be stored and distributed in the , or may be distributed via a communication network such as the Internet.

The present invention is capable of various embodiments and modifications without departing from the broader spirit and scope of the invention. Moreover, the above-described embodiments are for explaining the present invention, and do not limit the scope of the present invention. In other words, the scope of the present invention is indicated by the claims rather than the embodiments. Various modifications made within the scope of the claims and within the meaning of the invention equivalent thereto are considered to be within the scope of the present invention.

INDUSTRIAL APPLICATION This invention can be suitably employ|adopted for the air-conditioning control apparatus which controls an air conditioner.

100 air conditioning system 200 office space 201 window 202 wall 10 air conditioner 20 infrared camera 30 air conditioning control device N1 network 31 communication interface 32 processor 33 ROM 34 RAM 35 secondary storage device 36 Bus 351 Window detection program 352 Air conditioning control program 353 Image DB 354 Determination result DB 301 Acquisition unit 302 Window detection unit 303 Open/close determination unit 304 Air conditioning control unit 3021 Window determination execution unit 3022 Search area setting Department

Claims (9)

Acquisition means for acquiring thermal image data of a target region over a predetermined period;
window detection means for detecting a window region within the target region from a plurality of the thermal image data acquired by the acquisition means and having different acquisition times;
with
The window detection means is
window discrimination execution means for repeatedly executing a window discrimination process for discriminating whether or not each divided region obtained by dividing the search region included in the target region is a window region until a predetermined end condition is satisfied;
Search area setting means for setting the search area to be targeted in the next window discrimination process based on the discrimination result of the window discrimination process,
The window determination executing means includes, as the window determination process, a first window determination process and a second window determination process which is executed after the first window determination process and determines the window region by a method different from the first window determination process. Repeatedly executing the discrimination process and
Window detection device. The window determination executing means obtains an average temperature of the divided area from each of the plurality of thermal image data as the first window determination process, and determines that the divided area is a window area based on the obtained average temperature distribution. determine whether or not
The window detection device according to claim 1. As the second window discrimination processing, the window discrimination executing means obtains the average temperature of the divided regions from each of the plurality of thermal image data, and converts the obtained average temperature over time into a predetermined approximation curve. applying, determining whether the divided area is a window area based on the correlation between the fitted approximated curve and each average temperature;
The window detection device according to claim 1 or 2. the processing load of the second window discrimination process is greater than the processing load of the first window discrimination process;
The window detection device according to any one of claims 1 to 3. Further comprising open/close determination means for determining the open/closed state of the window area detected by the window detection means,
The window detection device according to any one of claims 1 to 4. a window detection device according to any one of claims 1 to 5;
air conditioning control means for controlling an air conditioner based on a detection result by the window detection means;
Air conditioning control device. An air conditioning system comprising: an imaging device that captures thermal image data of a target region; an air conditioner that air-conditions a region that includes the target region; and an air conditioning control device that controls the air conditioner,
The air conditioning control device
Acquisition means for acquiring thermal image data of a target region from the imaging device over a predetermined period;
window detection means for detecting a window region within the target region from a plurality of the thermal image data acquired by the acquisition means and having different acquisition times;
air conditioning control means for controlling the air conditioner based on the detection result of the window detection means;
The window detection means is
window discrimination execution means for repeatedly executing a window discrimination process for discriminating whether or not each divided region obtained by dividing the search region included in the target region is a window region until a predetermined end condition is satisfied;
Search area setting means for setting the search area to be targeted in the next window discrimination process based on the discrimination result of the window discrimination process,
The window determination executing means includes, as the window determination process, a first window determination process and a second window determination process which is executed after the first window determination process and determines the window region by a method different from the first window determination process. Repeatedly executing the discrimination process and
air conditioning system. an acquisition step of acquiring thermal image data of the region of interest over a predetermined period of time;
a window detection step of detecting a window region within the target region from a plurality of the thermal image data acquired at the acquisition step and having different acquisition times;
In the window detection step, the window region is detected by alternately and repeatedly executing a first window determination process and a second window determination process using different methods until a predetermined termination condition is satisfied .
Window detection method. the computer,
Acquisition means for acquiring thermal image data of the target area over a predetermined period;
window detection means for detecting a window region within the target region from a plurality of the thermal image data acquired by the acquisition means and having different acquisition times;
function as
The window detection means is
window discrimination execution means for repeatedly executing a window discrimination process for discriminating whether or not each divided region obtained by dividing the search region included in the target region is a window region until a predetermined end condition is satisfied;
Search area setting means for setting the search area to be targeted in the next window discrimination process based on the discrimination result of the window discrimination process,
The window determination executing means includes, as the window determination process, a first window determination process and a second window determination process which is executed after the first window determination process and determines the window region by a method different from the first window determination process. Repeatedly executing the discrimination process and
program.

Images