A pet waste identification method, system, and computer-readable storage medium

Abstract

The present application relates to the field of image recognition, and more particularly to a pet excrement recognition method, system and computer readable storage medium, the present application comprises: step one, through infrared thermal sensor to identify the heat signal generated by the temperature change in the environment, when the infrared thermal sensor detects the suspicious area, the suspicious area image is collected through the image collector;Step two, prepare the trained convolutional neural network, the convolutional neural network is trained based on pet excrement image, environmental obstacle image and thermal image under the obstacle, the convolutional neural network is used for inputting the infrared thermal image of suspicious area and the suspicious area image, judging the relative relationship of excrement, obstacle and heat flow generated by excrement, outputting whether there is pet excrement, pet excrement position, whether pet excrement is shielded by obstacle, pet excrement is located behind the obstacle probability.The technical scheme of the present application can effectively enhance the recognition effect by combining infrared thermal sensing with AI.

Description

Technical Field

[0001] This invention relates to the field of image recognition, and more specifically to a method, system, and computer-readable storage medium for identifying pet excrement. Background Technology

[0002] With the increasing number of pet owners, the timely disposal of pet excrement has become a critical issue in ensuring environmental hygiene. Traditional detection methods have many shortcomings, and simple infrared thermal detection may have a certain degree of false positives. While using artificial intelligence to assist in such detection can greatly improve accuracy, it requires suitable training data. Currently, there is a lack of a comprehensive solution that effectively combines infrared thermal sensing with AI and utilizes video stream training.

[0003] In existing technologies, for example, patent publication number CN108152238A discloses an intelligent pet excrement collection system based on infrared detection. This system identifies the temperature inside the container using infrared image recognition and then triggers the excrement collection action through image recognition. However, in this process, the infrared light is only used for temperature monitoring, and it is difficult to establish a correlation and enhancement between the infrared light and the image recognition function, resulting in limited recognition effectiveness. Summary of the Invention

[0004] To address the aforementioned issues, this invention provides a pet excrement identification method, system, and computer-readable storage medium that can effectively enhance identification performance by combining infrared thermal sensing with AI.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows: A method for identifying pet excrement, comprising:

[0006] Step 1: Identify the thermal signals generated by temperature changes in the environment using an infrared thermal sensor. Once the infrared thermal sensor detects a suspicious area, an image acquisition device is used to capture an image of the suspicious area.

[0007] Step 2: Prepare the training convolutional neural network. The convolutional neural network is trained based on images of pet excrement, images of obstacles in the environment, and thermal images of the obstacle environment. The convolutional neural network is used to input the infrared thermal image and image of the suspicious area, determine the relative relationship between excrement, obstacles, and the heat flow generated by the excrement, and output whether pet excrement exists, the location of pet excrement, whether pet excrement is occluded by obstacles, and the probability that pet excrement is located behind obstacles.

[0008] The above approach has the following beneficial effects:

[0009] 1. In this solution, an infrared thermal sensor is used to identify the heat signal generated by temperature changes in the environment. When a pet defecates, the infrared thermal sensor can detect the heat signal generated by the temperature change, thereby triggering the detection, so that the image detection process does not require constant monitoring of the image.

[0010] 2. In this solution, the convolutional neural network can be trained based on images of pet excrement, images of obstacles in the environment, and thermal images of the obstacle environment;

[0011] When pet excrement is exposed unobstructed in an image of a suspicious area, a convolutional neural network identifies the shape and color of the excrement to determine its presence and location.

[0012] When pet excrement is partially or completely obscured, the infrared thermal sensor has a certain degree of penetration and can image the area regardless of obstacles of a certain size or thickness. This helps to extract detailed features from the image of the suspicious area and identify the features of the pet excrement exposure through a convolutional neural network, thereby determining whether there is pet excrement, the location of the pet excrement, and whether the pet excrement is obscured by obstacles.

[0013] In addition to its penetrating ability, the infrared thermal sensor can also detect the rising heat flow from excrement, which is moist and warm after being discharged. This heat flow can be detected even when the excrement is completely blocked. The convolutional neural network can then identify obstacle information. When the position and volume of the obstacle match the conditions of covering the excrement or obscuring and guiding the heat flow, the output will indicate whether there is pet excrement, the location of the pet excrement, whether the pet excrement is blocked by the obstacle, and the probability that the pet excrement is behind the obstacle.

[0014] Furthermore, infrared thermal sensors and image acquisition devices are mobilely deployed around pet activity areas or near locations where pets frequently defecate.

[0015] Beneficial effects: Indoor environments are complex and contain many blind spots hidden by furniture. Therefore, deploying infrared thermal sensors and image acquisition devices around pet activity areas or near places where pets frequently defecate can achieve better detection results.

[0016] Furthermore, the thermal signal acquired by the infrared thermal sensor in step one is amplified by a preamplifier.

[0017] Beneficial effect: The amplifier enables infrared thermal sensors to be more precise, thus allowing for the detection of smaller temperature changes.

[0018] Furthermore, in step one, the infrared thermal sensor acquires the thermal signal and then performs noise reduction processing.

[0019] Beneficial effects: Noise reduction can reduce glitches in the infrared thermal images acquired by infrared thermal sensors, thereby improving the accuracy of judgment.

[0020] Furthermore, convolutional neural networks are also used to learn the patterns of pet excretion. Infrared thermal sensors identify pets within 2 to 4 hours of waking up and eating based on these patterns.

[0021] Beneficial effects: Pets exhibit regular excretion patterns, typically occurring within 2 to 4 hours of waking and eating. By learning these patterns, detection during low-probability periods can be reduced, thus minimizing ineffective monitoring and computational resource consumption.

[0022] Furthermore, the infrared thermal sensor and image acquisition device are based on learning the pet's excretion patterns and move to the vicinity of the pet's activity area or the location where the pet frequently excretes within 2 to 4 hours when the pet wakes up and eats.

[0023] Beneficial effects: Infrared thermal sensors and image acquisition devices can be deployed and integrated into mobile devices such as robotic vacuum cleaners. Based on learning the pet's defecation patterns, they can move to the vicinity of the pet's activity area or the vicinity of the pet's frequently defecated locations, thus facilitating dynamic deployment and improving convenience.

[0024] A pet excrement recognition system includes:

[0025] Sensor layer: includes infrared thermal sensors and video capture cameras;

[0026] Signal processing layer: Amplifies the thermal signal acquired by the infrared thermal sensor through a preamplifier, removes noise interference, and then converts the analog signal into a digital signal through an analog-to-digital converter; Video stream data acquired by the video capture camera is transmitted through a video capture card and encoded and compressed.

[0027] The computational processing layer includes a microprocessor, a graphics processing unit, and an access memory. The microprocessor is responsible for coordinating the work of each module, the graphics processing unit is used to provide parallel recognition and computation, and the access memory is used to temporarily store data and programs.

[0028] Beneficial effects: The infrared thermal sensor and video acquisition camera in the sensor layer can acquire infrared thermal images and images of suspicious areas; the signal processing layer provides preprocessing capabilities for infrared thermal images and images of suspicious areas; and the computational processing layer provides recognition and computation capabilities for images.

[0029] Furthermore, it also includes a communication and output layer, which is used to send the detection results to the user's mobile terminal.

[0030] Beneficial effect: After identifying excrement, the communication and output layer can send the detection results of the computing layer to the user's mobile terminal via remote communication, thereby alerting the user.

[0031] Furthermore, the communication and output layer is also equipped with an alarm output interface, which is connected to an audible and visual alarm. The audible and visual alarm is used to issue an alarm after the processing layer identifies the excrement.

[0032] Beneficial effects: The sound and light alarm can alert users by sound and flashing lights, prompting them to take timely action.

[0033] A computer-readable storage medium for pet excrement recognition includes a high-speed solid-state drive for storing a convolutional neural network model, training sample data, system programs, and detection results; and an external high-capacity hard drive for long-term storage of video stream data to provide training samples.

[0034] Beneficial effects: High-speed solid-state drives can store convolutional neural network models, training sample data, system programs, and detection results. External large-capacity hard drives can be used for low-cost storage of video stream data and facilitate the provision of samples for subsequent training.

[0035] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0036] Figure 1 This is a schematic diagram illustrating the steps of an embodiment of the pet excrement identification method of the present invention;

[0037] Figure 2 This is a schematic diagram of a module in an embodiment of the pet excrement recognition system of the present invention. Detailed Implementation

[0038] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0039] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0040] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0041] The following detailed description illustrates the specific implementation method:

[0042] As attached Figures 1-2 As shown: A method for identifying pet excrement, comprising:

[0043] Step 1: Identify the thermal signals generated by temperature changes in the environment using an infrared thermal sensor. Once the infrared thermal sensor detects a suspicious area, an image acquisition device is used to capture an image of the suspicious area.

[0044] Step 2: Prepare the training convolutional neural network. The convolutional neural network is trained based on images of pet excrement, images of obstacles in the environment, and thermal images of the obstacle environment. The convolutional neural network is used to input the infrared thermal image and image of the suspicious area, determine the relative relationship between excrement, obstacles, and the heat flow generated by the excrement, and output whether pet excrement exists, the location of pet excrement, whether pet excrement is occluded by obstacles, and the probability that pet excrement is located behind obstacles.

[0045] By using an infrared thermal sensor to identify the heat signal generated by temperature changes in the environment, when a pet defecates, the infrared thermal sensor can detect the heat signal generated by the temperature change, thereby triggering the detection. This eliminates the need for constant monitoring of the image during the image detection process, which is triggered only when a temperature fluctuation is detected, thus saving normal computing resources.

[0046] Since infrared thermal sensors can only acquire infrared thermal images, which only contain temperature and general shape information and lose details such as texture and color of pet excrement, they need to be combined with image recognition for more accurate detection.

[0047] Convolutional neural networks can be trained based on images of pet excrement, images of obstacles in the environment, and thermal images of obstacles.

[0048] When pet excrement is exposed unobstructed in an image of a suspicious area, a convolutional neural network identifies details such as the shape and color of the excrement to determine whether pet excrement exists and its location.

[0049] However, pets don't only defecate in open areas. They often relieve themselves in corners or near pillars, where there may be obstructions. They may also defecate in pit-like spaces, completely obscuring their excrement. When pet excrement is partially or completely obscured, the infrared thermal sensor's infrared detection has a certain penetrating power, allowing it to image even through obstacles of a certain size or thickness. This aids in extracting detailed features from images of suspicious areas. Furthermore, a convolutional neural network identifies the exposed features of pet excrement, thus determining the presence, location, and obstruction of the excrement.

[0050] In addition to its penetrating ability, the infrared thermal sensor can also detect the rising heat flow from excrement, which is moist and warm after being discharged. This heat flow can be detected even when the excrement is completely blocked. The convolutional neural network can then identify obstacle information. When the position and volume of the obstacle match the conditions of covering the excrement or obscuring and guiding the heat flow, the output will indicate whether there is pet excrement, the location of the pet excrement, whether the pet excrement is blocked by the obstacle, and the probability that the pet excrement is behind the obstacle.

[0051] In step one, the thermal signal acquired by the infrared thermal sensor is amplified by a preamplifier.

[0052] In step one, the infrared thermal sensor acquires the thermal signal and then performs noise reduction processing.

[0053] Amplifiers enable infrared thermal sensors to capture finer details, allowing for the detection of smaller temperature variations. Noise reduction reduces glitches in the infrared thermal images acquired by the sensor, improving the accuracy of judgments.

[0054] Convolutional neural networks are also used to learn the patterns of pet excretion. Infrared thermal sensors can identify pets within 2 to 4 hours of waking up and eating based on these patterns.

[0055] Pets exhibit regular elimination patterns, typically occurring within 2 to 4 hours of waking and eating. By learning these patterns, detection during low-probability periods can be reduced, minimizing unnecessary listening and computational resource consumption. Infrared thermal sensors and image acquisition devices can be integrated into mobile devices such as robotic vacuum cleaners, moving to areas surrounding pet activity or frequently defecating locations based on these patterns, facilitating dynamic deployment and improving convenience.

[0056] A pet excrement recognition system includes:

[0057] Sensor layer: includes infrared thermal sensors and video capture cameras;

[0058] Signal processing layer: Amplifies the thermal signal acquired by the infrared thermal sensor through a preamplifier, removes noise interference, and then converts the analog signal into a digital signal through an analog-to-digital converter; Video stream data acquired by the video capture camera is transmitted through a video capture card and encoded and compressed.

[0059] The computational processing layer includes a microprocessor, a graphics processing unit, and an access memory. The microprocessor is responsible for coordinating the work of each module, the graphics processing unit is used to provide parallel recognition and computation, and the access memory is used to temporarily store data and programs.

[0060] Communication and Output Layer: The communication and output layer is used to send the detection results to the user's mobile terminal. The communication and output layer also has an alarm output interface, which is connected to an audible and visual alarm. The audible and visual alarm is used to sound an alarm after the processing layer identifies excrement.

[0061] The infrared thermal sensor and video capture camera in the sensor layer can acquire infrared thermal images and images of suspicious areas. The signal processing layer provides preprocessing capabilities for infrared thermal images and images of suspicious areas, and the computational processing layer provides recognition and computation capabilities for images.

[0062] The communication and output layer is also equipped with an alarm output interface, which is connected to an audible and visual alarm. The audible and visual alarm is used to issue an alarm after the excrement is identified in the computing layer.

[0063] Audible and visual alarms can alert users by sound and flashing lights, prompting them to take timely action.

[0064] A computer-readable storage medium for pet excrement recognition includes a high-speed solid-state drive for storing a convolutional neural network model, training sample data, system programs, and detection results; and an external high-capacity hard drive for long-term storage of video stream data to provide training samples.

[0065] High-speed solid-state drives can store convolutional neural network models, training sample data, system programs, and detection results. External large-capacity hard drives are used for low-cost storage of video stream data and facilitate the provision of samples for subsequent training.

[0066] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for identifying pet excrement, characterized in that, include: Step 1: Identify the thermal signals generated by temperature changes in the environment using an infrared thermal sensor. Once the infrared thermal sensor detects a suspicious area, an image acquisition device is used to capture an image of the suspicious area. Step 2: Prepare the training convolutional neural network. The convolutional neural network is trained based on images of pet excrement, images of obstacles in the environment, and thermal images of the obstacle environment. The convolutional neural network is used to input the infrared thermal image and image of the suspicious area, determine the relative relationship between excrement, obstacles, and the heat flow generated by the excrement, and output whether pet excrement exists, the location of pet excrement, whether pet excrement is occluded by obstacles, and the probability that pet excrement is located behind obstacles.

2. The pet excrement identification method according to claim 1, characterized in that, Infrared thermal sensors and image acquisition devices are mobilely deployed around pet activity areas or near locations where pets frequently defecate.

3. The pet excrement identification method according to claim 2, characterized in that, In step one, the thermal signal acquired by the infrared thermal sensor is amplified by a preamplifier.

4. The pet excrement identification method according to claim 3, characterized in that, In step one, the infrared thermal sensor acquires the thermal signal and then performs noise reduction processing.

5. The pet excrement identification method according to claim 4, characterized in that, Convolutional neural networks are also used to learn the patterns of pet excretion. Infrared thermal sensors can identify pets within 2 to 4 hours of waking up and eating based on these patterns.

6. The pet excrement identification method according to claim 5, characterized in that, Infrared thermal sensors and image acquisition devices are based on learning the pet's excretion patterns and move to the vicinity of the pet's activity area or the location where the pet frequently excretes within 2 to 4 hours when the pet wakes up and eats.

7. A pet excrement identification system, based on the pet excrement identification method of claim 6, characterized in that, include: Sensor layer: includes infrared thermal sensors and video capture cameras; Signal processing layer: Amplifies the thermal signal acquired by the infrared thermal sensor through a preamplifier, removes noise interference, and then converts the analog signal into a digital signal through an analog-to-digital converter; Video stream data acquired by the video capture camera is transmitted through a video capture card and encoded and compressed. The computational processing layer includes a microprocessor, a graphics processing unit, and an access memory. The microprocessor is responsible for coordinating the work of each module, the graphics processing unit is used to provide parallel recognition and computation, and the access memory is used to temporarily store data and programs.

8. The pet excrement identification system according to claim 7, characterized in that, It also includes a communication and output layer, which is used to send the detection results to the user's mobile terminal.

9. The pet excrement identification system according to claim 8, characterized in that, The communication and output layer is also equipped with an alarm output interface, which is connected to an audible and visual alarm. The audible and visual alarm is used to issue an alarm after the excrement is identified in the computing layer.

10. A computer-readable storage medium for pet excrement identification, based on the computer-readable storage medium of the pet excrement identification system of claim 9, characterized in that, It includes a high-speed solid-state drive for storing convolutional neural network models, training sample data, system programs, and detection results; it also includes an external high-capacity hard drive for long-term storage of video stream data to provide training samples.

Images