Droplet detection device
The droplet detection device enhances detection accuracy by using air blowing and thermal imaging to create temperature differences, addressing distance and thermal equilibrium challenges in existing technologies.
Patent Information
- Application Number
- JP2023208940
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-12-12
- Publication Date
- 2025-06-24
AI Technical Summary
Existing droplet detection technologies face challenges in accurately detecting droplets at varying distances from the camera and when droplets and their surroundings reach thermal equilibrium, leading to difficulty in distinguishing them in thermal images.
A droplet detection device that includes air blowing means to create a temperature difference between droplets and the floor surface, using a thermal image capturing means to capture temperature distribution, and droplet detection means to identify droplets based on temperature differences.
Improves droplet detection accuracy by creating a temperature contrast between droplets and the floor, enabling detection even when they are colorless, transparent, or small, and reduces distance-related detection issues.
Smart Images

Figure 2025093351000001_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a droplet detection device.
Background Art
[0002] In stores such as convenience stores and supermarkets, on the floor surfaces of facilities, and in toilets, there are situations where droplets adhere to the floor surfaces. On the toilet floor, urine droplets may scatter, or water may scatter during handwashing. If left unattended, there are problems such as being unhygienic and generating odors. Also, on the floor surfaces of stores such as convenience stores and supermarkets and facilities, beverages may spill, or drops from wet umbrellas in the rain may wet the store floor. If left unattended, there are problems such as users slipping and getting injured, soiling the users' shoes, and being unhygienic. Regarding the detection of droplets, the techniques described in the following Non-Patent Documents 1 to 3 are known.
[0003] Non-Patent Document 1 describes a technique for detecting water droplets adhering to a belt by performing machine learning on a plurality of images of the belt of a belt conveyor using AI (artificial intelligence). Non-Patent Document 2 describes a technique for detecting raindrops and the like adhering to a window from a video captured by a moving vehicle such as an automobile by image analysis and generating a video from which the raindrops are removed. Non-Patent Document 3 describes a technique for detecting water leakage from a pipe buried underground using a thermal image (infrared image).
Prior Art Documents
Non-Patent Documents
[0004]
Non-Patent Document 1
Non-Patent Document 2
Non-Patent Document 3
Summary of the Invention
Problems to be Solved by the Invention
[0005] (Problems of the Prior Art) In the prior art described in Non-Patent Documents 1 and 2, when the distance between the droplet and the camera is close, the droplet is detected by detecting the droplet-shaped noise generated on the image. However, when the droplet and the camera are far apart, there is a problem that it is difficult to detect. In the technology described in Non-Patent Document 3, when water is continuously leaking, the temperature difference between the water and the surroundings is likely to continuously occur, and it is easy to distinguish the water leakage and pipes, etc. in the thermal image. However, for the droplets scattered on the floor surface, the temperature of the droplets and the floor surface easily become the same temperature quickly, and there is a problem that it is difficult to detect in the thermal image.
[0006] The technical problem of the present invention is to improve the detection accuracy of droplets compared with the prior art.
Means for Solving the Problem
[0007] In order to solve the above technical problem, the droplet detection device of the invention according to claim 1 comprises blowing means for blowing air toward a detection target part that is a target for detecting droplets, thermal image capturing means for capturing a thermal image of the detection target part blown by the blowing means, and droplet detection means for detecting droplets based on the temperature distribution in the thermal image. It is characterized by being provided with the above.
Advantages of the Invention
[0008] According to the invention described in claim 1, the detection accuracy of droplets can be improved as compared with the prior art.
Brief Description of the Drawings
[0009]
Figure 1
Figure 2
Figure 3
Figure 4
Modes for Carrying Out the Invention
[0010] Next, specific examples of embodiments of the present invention (hereinafter referred to as embodiments) will be described with reference to the drawings, but the present invention is not limited to the following embodiments. In the following description using the drawings, illustrations other than the members necessary for the explanation are appropriately omitted for ease of understanding.
Embodiment
[0011] FIG. 1 is an explanatory diagram of the droplet detection device according to Embodiment 1 of the present invention. In FIG. 1, in the individual toilet 2 where the droplet detection device 1 of Example 1 of the present invention is installed, the toilet bowl 9 is supported on the back wall 3. The toilet bowl 9 of Example 1 is in a so-called cantilever state, wall-mounted and supported on the back wall 3. Therefore, a gap is formed between the lower end of the toilet bowl 9 of Example 1 and the floor surface 8. On the floor surface 8, which is an example of the detection target part, a fan 11, which is an example of the air blowing means, is installed. The fan 11 blows air against the floor surface 8. Note that the air blowing direction of the fan 11 is set in the horizontal direction along the floor surface 8, but it is also possible to set it downward from the horizontal direction, and the air blowing direction can be arbitrarily changed as long as air can be blown onto the floor surface 8. In Example 1, a plurality of fans 11 are arranged side by side along one side of the outer edge of the floor surface 8. The fans 11 of Example 1 are installed along the outer edge of the floor surface 8 without gaps so as to be able to blow air over the entire area of the floor surface 8, but it is not limited to this. It is also possible to install the fans 11 with gaps along the outer edge.
[0012] Above the toilet 2, a thermal camera 12, which is an example of the thermal image capturing means, is installed. The thermal camera 12 of Example 1 is a camera that captures a thermal image, that is, a temperature distribution as an image, and a conventionally known and commercially available infrared thermal camera can be preferably used. Also, in the toilet 2, a wiping cleaning robot 13, which is an example of the cleaning means, is arranged. When an instruction for cleaning is given, the wiping cleaning robot 13 autonomously travels to wipe the floor surface 8. The fan 11 and the thermal camera 12 are connected to the control means (controller) C. Also, the wiping cleaning robot 13 is configured to be able to receive a control signal wirelessly (wireless communication possible) from the control means C.
[0013] FIG. 2 is a functional block diagram of the control unit of Example 1. (Description of the control circuit) The control means C of Example 1 is composed of an I / O (input / output interface) that performs input / output of signals with the outside and adjustment of input / output signal levels, a ROM (read-only memory) that stores programs and data for performing necessary startup processing, a RAM (random access memory) that temporarily stores necessary data and programs, a CPU (central processing unit) that performs processing according to the startup program stored in the ROM, etc., and a microcomputer device having a clock oscillator, etc. The control means C can realize various functions by executing the programs stored in the ROM, RAM, etc.
[0014] The control means C of Example 1 receives signals from signal output elements such as the thermal camera 12 and sensors (not shown). The control means C of Example 1 outputs control signals to controlled elements such as the fan 11 and the wiping robot 13. The droplet detection program of the control means C of Example 1 has the following functions (means, functional modules) C1 to C5.
[0015] The detection start determination means C1 determines whether it is time to detect droplets on the floor surface 8. The detection start determination means C1 of Example 1 determines that it is time to detect droplets when the current time reaches a predetermined time. Note that the droplet detection time is not limited to a specific time. For example, it is possible that the user of the toilet 2 has left the room. That is, by providing a sensor that detects the opening and closing of the door of the toilet 2 or a human sensor that detects the presence or absence of a person in the toilet 2, etc., it is also possible to determine that it is time to detect droplets when the use of the toilet 2 is detected. Note that it is also possible to determine that it is time to detect droplets every time the toilet 2 is used, or it is also possible to determine that it is time to detect droplets when the use of the toilet 2 reaches a predetermined number of times.
[0016] The air supply control means C2 controls the fan 11 to control the air supply to the floor surface 8 and the stop of the air supply. The air supply control means C2 of the first embodiment starts the air supply to the floor surface 8 when it is time to detect droplets, and stops the air supply when the acquisition of the thermal image is completed. Note that the end time of the air supply is not limited to the completion of the acquisition of the thermal image, and can be appropriately changed according to design, use, etc., such as an aspect where the air supply is ended when the detection of the droplets is completed. The thermal image acquisition means C3 acquires the thermal image taken by the thermal camera 12. The thermal image acquisition means C3 of the first embodiment acquires the thermal image of the floor surface 8 after the start of the air supply. Although it is preferable to use the thermal image during the air supply, it is also possible to acquire the thermal image in a state where the temperature of the droplets has decreased due to the latent heat of vaporization immediately after the air supply stops (a state before the temperature of the droplets reaches the temperature of the floor surface 8). In this case, the air supply ends before the acquisition of the thermal image.
[0017] FIG. 3 is an explanatory diagram of an example of a thermal image in which droplets are detected. The droplet detection means C4 detects droplets based on the thermal image acquired by the thermal image acquisition means C3. The droplet detection means C4 of the first embodiment detects, as the droplet 21, a portion having a temperature equal to or lower than a predetermined threshold value Ta with respect to the average temperature T1 of the thermal image. That is, the portion where the temperature T0 does not reach T1 - Ta is detected as the droplet 21. The droplet detection means C4 of the first embodiment detects the size (low-temperature range) and the position on the floor surface 8 of the droplet 21.
[0018] When the droplet 21 is detected on the floor surface 8, the wiping control means C5 transmits a control signal to the wiping robot 13 to perform wiping. The wiping control means C5 of the first embodiment transmits the control signal for instructing wiping to the wiping robot 13, including the position information of the droplet 21. Therefore, the wiping robot 13 moves to the position of the droplet 21 and performs wiping.
[0019] (Explanation of the flowchart of the first embodiment) Next, the control flow in the droplet detection device of the first embodiment will be described using a flowchart, so-called flow chart.
[0020] (Explanation of the flowchart) FIG. 4 is an explanatory diagram of the flowchart of the droplet detection program of Example 1. The processing of each step ST in the flowchart of FIG. 4 is performed according to the program stored in the control means C. Further, this processing is executed in parallel with various other processes of the control means C. The flowchart shown in FIG. 4 starts when the droplet detection program is activated in the control means C.
[0021] In ST1 of FIG. 4, it is determined whether it is the droplet detection time. If yes (Y), the process proceeds to ST2; if no (N), ST1 is repeated. In ST2, the fan 11 is operated. Then, the process proceeds to ST3. In ST3, a thermal image is acquired. Then, the process proceeds to ST4. In ST4, the fan 11 is stopped. Then, the process proceeds to ST5. In ST5, the droplet 21 is detected based on the thermal image. Then, the process proceeds to ST6. In ST6, it is determined whether the droplet 21 has been detected. If yes (Y), the process proceeds to ST7; if no (N), the process returns to ST1. In ST7, a control signal for instructing the wiping robot 13 to perform wiping is transmitted. Then, the process returns to ST1.
[0022] The droplet detection device 1 of Example 1 having the above configuration operates the fan 11 when it is the droplet detection time to acquire a thermal image. When the droplet 21 exists on the floor surface 8, when blown by the fan 11, in the process of evaporation of the solvent (mainly water) of the droplet 21, the temperature of the droplet 21 decreases due to the heat of vaporization. Therefore, even if the temperature of the droplet 21 and the temperature of the floor surface 8 are the same temperature, a temperature difference is generated between the droplet 21 and the floor surface 8 by the blowing of the fan 11. Using this temperature difference, the presence and position of the droplet 21 are detected in the thermal image. Therefore, even when the droplet 21 is colorless, transparent, or small, or when the temperatures of the droplet 21 and the floor surface 8 are the same, it is possible to detect the droplet 21. Further, in the first embodiment, the thermal camera 12 is used, and the distance between the camera and the droplet as in Non-Patent Documents 1 and 2 is less likely to be a problem. Therefore, the detection accuracy of the droplet 21 can be improved as compared with the conventional technologies such as Non-Patent Documents 1 to 3.
[0023] (Modification example) As described above, the embodiments of the present invention have been described in detail. However, the present invention is not limited to the above embodiments, and various modifications can be made within the scope of the gist of the present invention described in the claims. Modification examples (H01) to (H04) of the present invention are illustrated below. (H01) In the above embodiment, the floor surface 8 of the toilet 2 is illustrated as an example of the detection target portion, but the present invention is not limited thereto. It is applicable to the floor surfaces of stores such as convenience stores and supermarkets, and facilities such as exhibition halls and stations. In addition, the present invention is not limited to the floor surface, and is also applicable to desks, tables, etc., and belt-shaped objects such as belt conveyors and handrails of escalators.
[0024] (H02) In the above embodiment, it is preferably applied to the toilet 2 in which the wall-mounted toilet 9 is installed, but the present invention is not limited thereto. It is also applicable to toilets installed on the floor surface 8. (H03) In the above embodiment, the mode of operating the wiping robot 13 when the droplet 21 is detected is illustrated, but the present invention is not limited thereto. For example, it is also possible to notify and report to the administrator of the toilet 2 by notification means such as a lamp, a buzzer, voice guidance, and a display, and prompt cleaning. (H04) In the above embodiment, the form of operating the wiping robot 13 when the droplet 21 is detected is illustrated, but the present invention is not limited thereto. For example, every time a user of the toilet 2 leaves the room, the detection of the droplet 21 is performed each time and stored as history information. When a specific cleaning time arrives, cleaning is performed on the location where the droplet 21 was detected between the previous cleaning and the current cleaning based on the history information.
Explanation of reference numerals
[0025] 1…Droplet detection device, 8…Detection target part, floor surface, 11…Air blowing means, 12…Thermal imaging means, 21…Droplet, C4…Droplet detection means.
Claims
【Claim 1】 Blowing means for blowing air toward a detection target portion that is a target for detecting droplets; Thermal image capturing means for capturing a thermal image of the detection target portion blown by the blowing means; Droplet detection means for detecting droplets based on the temperature distribution in the thermal image; A droplet detection device, characterized by comprising the above.