Pet recognition based feeding system and method

Abstract

The application provides a pet-identification-based feeding system and method, comprising a feeder; the feeder comprises a master control unit, an image sensor, a feeding trough and a weight sensor; the weight sensor collects weight data of the feeding trough; the image sensor collects image data of a feeding pet; and the master control unit acquires feeding data of the pet identified based on the image data according to the weight data. The application collects identity information of the feeding pet through the image sensor, measures the feeding amount of the pet through the weight sensor, solves the defect that the existing pet feeder cannot acquire the feeding amount of each pet, and solves the discomfort of the pet when wearing a collar compared with acquiring the identity information by wearing a collar.

Description

Technical Field

[0001] This invention relates to the field of pet feeding device technology, and more specifically, to a feeding system and method based on pet identification. Background Technology

[0002] With socio-economic development and the application of intelligent products, people are pursuing spiritual fulfillment and enjoying the companionship of pets such as cats and dogs. The demand for pets is increasing daily, resulting in a large market for pet products. Simultaneously, people's awareness of scientific pet care is gradually improving, seeking healthy and reasonable pet-raising lifestyles and avoiding problems such as irregular feeding due to insufficient time and energy. This shift in pet-raising awareness, when transmitted to the production end, will bring new opportunities to the industry of scientifically designed and intelligent feeding products.

[0003] When there is only one pet, a traditional pet feeder can ensure that the pet eats regularly by dispensing food at set times and in measured amounts. However, when there are more than one pet, a traditional feeder cannot obtain feeding data for each pet.

[0004] Chinese patent document CN212232614U discloses a feeding device with multi-cat identification function, including a feeder body, an electronic tag, and a base fixed to the lower end of the feeder body. The base houses a signal receiver and a battery, which are electrically connected. The base also houses a controller, which is electrically connected to both the signal receiver and the battery. The electronic tag works in conjunction with the signal receiver. By attaching the electronic tag to the cat's body, when the cat approaches the feeder body to eat, the signal receiver receives the radio frequency signal from the electronic tag to identify the cat. However, this solution requires each cat to wear its own unique identification collar, increasing user costs and affecting the cat's comfort. Summary of the Invention

[0005] In view of the deficiencies in the prior art, the purpose of this invention is to provide a feeding system and method based on pet identification.

[0006] A pet-recognition-based feeding system according to the present invention includes a feeder;

[0007] The feeder includes a main control unit, an image sensor, a feeding trough, and a weight sensor;

[0008] The weight sensor collects weight data from the feeding trough;

[0009] The image sensor collects image data of the pet eating;

[0010] The main control unit acquires the pet's feeding data based on the image data, according to the weight data.

[0011] Preferably, the feeding trough is provided with enclosure mechanisms on both sides, and the enclosure mechanisms on both sides form openings;

[0012] The main control unit obtains the feeding data of a single pet that has entered the enclosure mechanism based on the image data, according to the weight data.

[0013] Preferably, the feeding data includes the weight of food consumed;

[0014] The weight of food consumed was obtained in the following manner:

[0015] The main control unit determines whether there is a pet in the target area using a proximity sensor. If there is no pet, the main control unit controls the weight sensor to enter standby mode. If there is a pet, the main control unit uses a weighing sensor to measure the weight of the feeding trough when the pet enters the target area and when the pet leaves the target area. The feeding weight is the difference between the weight of the feeding trough and the weight of the feeding trough.

[0016] Preferably, the feeding data includes the number of times a meal is eaten;

[0017] The number of meals was obtained in the following way:

[0018] The main control unit uses a proximity sensor to determine if there is a pet in the target area; if there is no pet, the main control unit controls the image sensor and weight sensor to enter standby mode; if there is a pet, the main control unit uses a weighing sensor to determine if the weight data of the feeding trough has changed.

[0019] If the weight data does not change, the number of meals is not counted; if the weight data changes, the number of meals is counted.

[0020] Preferably, the feeding data includes the duration of feeding;

[0021] The duration of the meal was obtained in the following way:

[0022] The main control unit uses a proximity sensor to determine if there is a pet in the target area; if there is no pet, the main control unit controls the image sensor and weight sensor to enter standby mode; if there is a pet, the main control unit uses a weighing sensor to determine if the weight data of the feeding trough has changed.

[0023] If the weight data does not change, the current time is not included in the feeding time, and the image sensor remains in standby mode; if the weight data changes, the main control unit uses the image sensor to identify the pet's head posture.

[0024] If the pet's head is lowered, the current time is counted towards the feeding time; if the pet's head is not lowered, the current time is not counted towards the feeding time.

[0025] Alternatively, the duration of the meal can be obtained in the following way:

[0026] The main control unit uses a proximity sensor to determine if there is a pet in the target area; if there is no pet, the main control unit controls the image sensor and weight sensor to enter standby mode; if there is a pet, the main control unit uses a weighing sensor to determine if the weight data of the feeding trough has changed.

[0027] If the weight data does not change, the current time is not included in the feeding time, and the image sensor remains in standby mode; if the weight data changes, the period of time during which the weight data changes is the feeding time.

[0028] Preferably, the image data is analyzed through a pet identification network to obtain the corresponding pet ID, and the main control unit associates the pet ID with the feeding data.

[0029] Preferably, the food in the feeding trough is delivered in any of the following ways;

[0030] Method 1: The main control unit collects the weight data of the feeding trough through a weighing sensor. When the weight data is lower than the threshold, the main control unit controls the feeding of food until the weight of the feeding trough reaches the maximum set value.

[0031] Method 2: Set a single feeding amount for each pet. The main control unit feeds the pets identified based on image data in a single feeding. The single feeding amount is the single feeding amount minus the remaining food in the feeding trough.

[0032] Preferably, the weight data is collected when the feeding trough is in a stable state, wherein the weight of the feeding trough does not change within a set time period.

[0033] Preferably, the automatic feeding device includes a business logic server;

[0034] The business logic server is used to store pets' feeding data and pet IDs, and displays them by connecting to user terminal devices.

[0035] Preferably, the automatic feeding device further includes an image server;

[0036] The business logic server obtains the pet ID and feeding data in the following ways;

[0037] The image sensor uploads image data to the image server and generates an image ID corresponding to the image data. The main control unit uploads the feeding data and the image ID to the business logic server.

[0038] The logic server retrieves image data from the image server based on the image ID, obtains the pet ID corresponding to the image data through the pet identification network, and associates the pet ID with the feeding data.

[0039] According to a pet feeding method provided by the present invention, the pet-based feeding system is used for feeding.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] 1. This invention collects the pet's identity information by using an image sensor and measures the pet's food intake and other feeding data by using a weight sensor. This solves the problem that existing pet feeders cannot obtain feeding data for each pet. Compared with using a collar to obtain identity information, it also solves the discomfort that pets experience when wearing the collar.

[0042] 2. The enclosure mechanism of this invention ensures that only one pet can eat at a time, avoiding inaccurate statistics caused by multiple pets eating at the same time.

[0043] 3. The feeding quantity statistics method designed in this invention only activates the image sensor when the proximity sensor acquires data and the weight of the feeding trough changes, which effectively reduces the power consumption caused by the long-term operation of the image sensor, thereby improving the battery life of the automatic feeding device.

[0044] 4. This invention sets the image data storage and recognition functions in the image server, which reduces the image computing power requirements of the feeder and lowers the overall manufacturing cost.

[0045] 5. This invention can connect to an image sensor via a user terminal, allowing owners to check their pets' living status at any time, including whether they are active or exhibiting any abnormalities, thereby enabling them to promptly detect health problems in their pets.

[0046] 6. The image sensor in this invention can also enable owners and pets to interact and have fun, enhancing the interaction and connection between owners and pets.

[0047] 7. In this invention, the weight sensor and the image sensor work together to count the feeding time only when the weight of the feeding trough changes and the pet is in a head-down posture. The feeding time is not counted when the pet is playing with its head down or just fiddling with the feeding trough, thus making the measurement results more accurate. In addition, the image sensor is activated only when the weight of the feeding trough changes, reducing the number of times the image sensor is activated, thereby improving the battery life of the automatic feeding device.

[0048] Other beneficial effects of the present invention will be explained in detail through the introduction of specific technical features and technical solutions in specific embodiments. Those skilled in the art should be able to understand the beneficial technical effects brought about by these technical features and technical solutions through the introduction of these technical features and technical solutions. Attached Figure Description

[0049] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0050] Figure 1 This is a schematic diagram of the automatic feeding device in this invention;

[0051] Figure 2 This is a topological diagram of the automatic feeding device in this invention;

[0052] Figure 3 This is a schematic diagram of the automatic feeding device with a enclosure mechanism in this invention.

[0053] Explanation of reference numerals in the attached figures:

[0054] Detailed Implementation

[0055] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0056] This invention discloses a feeding system based on pet identification, which can collect feeding data such as the identity information, amount of food consumed, number of feedings, and feeding time of each feeding pet, allowing owners to keep track of each pet's feeding situation and thus promptly detect the pet's health status.

[0057] Specifically, refer to Figure 1 and Figure 3 As shown, the automatic feeding device includes: a base 2 and a feeder body 1 fixedly mounted on the base 2; the feeder body 1 has a food bin, and a food bin cover 3 is provided at the top of the feeder body 1. Pet food is added from the top of the feeder. The feeder body 1 includes a feeder outlet 4. A feeding trough 6 is provided on the base 2 at the position corresponding to the feeder outlet 4. Enclosure mechanisms 7 are provided on both sides of the feeding trough 6, and the enclosure mechanisms 7 on both sides form openings 8. The enclosure mechanisms 7 are used to limit the number of pets eating at the same time, avoiding inaccurate statistics caused by multiple pets eating at the same time.

[0058] A weight sensor is installed on the base 2 to collect weight data of the feeding trough 6. An image sensor 5 is installed on the feeder body 1 to collect image data of the pet eating. In this embodiment, the image sensor 5 can be a camera, an infrared sensor, or other sensors capable of capturing images of the pet's identity. The automatic feeding device also includes a main control unit, and the weight sensor and image sensor 5 are electrically connected to the main control unit. When the pet is eating, the image sensor 5 collects image data of the pet located between the two side enclosure mechanisms 7. By analyzing and identifying the image data, it is determined which pet is eating. The weight of the feeding trough 6 is measured before and after the pet eats to obtain the amount of food consumed by the pet. The time when the weight of the feeding trough 6 changes is the pet's eating time.

[0059] The automatic feeding device also includes a proximity sensor, which is electrically connected to the main control unit. The proximity sensor detects whether a pet is approaching the automatic feeding device. In this embodiment, the proximity sensor can be configured to detect whether a pet is approaching the opening 8 of the enclosure mechanism 7. This feeding device only allows the image sensor 5 and weight sensor to operate when the proximity sensor acquires pet information, preventing frequent sensor activation and thus affecting the overall battery life of the automatic feeder.

[0060] Automatic feeding devices track pet food intake in the following ways:

[0061] Step S1: The main control unit acquires proximity sensor data in real time to determine whether there is a pet around the automatic feeding device. If not, the main control unit sends a standby command to the image sensor 5 and the weight sensor. If there is a pet, the weight sensor starts acquiring the weight data of the feeding trough 6.

[0062] Step S2: Determine if the weight of the feeding trough 6 has changed. If it has not changed, the main control unit sends a standby command to the image sensor 5. If it has changed, the image sensor 5 begins to collect pet image data. When the proximity sensor detects a pet at the opening 8 and the weight of the feeding trough 6 remains unchanged, it indicates that the pet is playing within the detection range of the proximity sensor, rather than eating. Therefore, the image sensor 5 remains in standby mode. Only when both the proximity sensor and the weight sensor collect data simultaneously, indicating that the pet is eating, will the image sensor 5 begin collecting data.

[0063] Since the image sensor 5 consumes a lot of power when it is working, the feeding statistics method disclosed in this invention can greatly reduce the working frequency of the image sensor 5 and improve the overall battery life, which is especially effective for rechargeable feeders.

[0064] Step S3: Acquire the pet's identity and posture data via image sensor 5. If the pet's posture is a head-down posture, it is determined to be in a "feeding state," and the main control unit starts timing to calculate the amount of food consumed in this state. If the pet's posture is not a head-down posture, it is determined to be in a "non-feeding state," and timing stops. When the pet is playing, its feeding process is not continuous. For example, it may eat with its head down for 1 minute, then raise its head to play or rest for 2 minutes. During this time, it may touch the feeding trough 6, causing a change in the weight of the trough 6. In this case, only the time spent in the head-down posture is recorded to obtain a more accurate feeding time. The pet's feeding time can reflect its state while eating. For example, when a pet likes a particular food, its appetite will increase, or when the pet has a physical abnormality, such as a toothache, the feeding time will increase. Analyzing the feeding time makes it easier to detect abnormalities in the pet's body in a timely manner.

[0065] Step S4: The main control unit stores the pet's identity information, feeding time, and food intake data accordingly. This allows owners to easily track each pet's feeding time and amount. The main control unit can also collect the number of times a pet eats and store the feeding frequency data along with the pet's identity information.

[0066] The method for replenishing feed in the feeding trough 6 includes: 1. When the weight of the feed is below a threshold, feed is added to the feeding trough 6 through the feed outlet 4 of the feeder until the weight reaches the set value. 2. The pet's identity information is obtained through the image sensor 5, and feed is dispensed according to the preset feeding amount for each pet. The dispensing amount = the required feeding amount - the remaining amount of feed in the feeding trough 6. The first dispensing method can reduce food waste, while the second dispensing method can control the feeding amount for each pet, and can be flexibly selected according to needs.

[0067] In a preferred embodiment, the automatic feeding device further includes an image server 9 and a business logic server 10 for reference. Figure 2 As shown, the automatic feeding device operates as follows:

[0068] When a pet approaches the automatic feeder, the proximity sensor signal triggers the main control unit. Upon receiving the trigger signal, the main control unit instructs the image sensor 5 to start capturing images and uploads the collected data to the image server 9. If the upload is successful, the camera returns the image ID information to the main control unit. If the upload fails due to timeout, the retransmission mechanism is triggered until successful.

[0069] The business logic server 10 is electrically connected to the main control unit, acquires image ID data, feeding time data, and food intake data, retrieves image data from the image server 9 based on the image ID data, performs machine deep learning through a pet identification network, and outputs pet identification information to the business logic server 10. The business logic server 10 statistically analyzes the pet identification information, feeding time, and food intake, and sends the statistical results to the user terminal.

[0070] The owner can also send control commands to the business logic server 10 through the user terminal to control the working status of the image sensor 5. For example, the owner can observe the pet's mental state and eating habits at home at any time, and can also interact with the pet through the image sensor 5.

[0071] The present invention will now be described in more detail through more preferred examples.

[0072] This invention discloses a pet-based feeding system, including a feeder; the feeder includes a main control unit, an image sensor 5, a feeding trough 6, and a weight sensor; the weight sensor collects weight data of the feeding trough 6; the image sensor 5 collects image data of the pet eating; the main control unit obtains the pet's eating data based on the weight data and the image data. The pet can be a cat or a dog.

[0073] Reference Figure 1 and Figure 3 As shown, the automatic feeding device includes: a base 2 and a feeder body 1 fixedly mounted on the base 2; the feeder body 1 has a food bin, and a food bin cover 3 is provided at the top of the feeder body 1. Pet food is added from the top of the feeder. The feeder body 1 includes a feeder outlet 4. A feeding trough 6 is provided on the base 2 at the position corresponding to the feeder outlet 4. Enclosure mechanisms 7 are provided on both sides of the feeding trough 6, and the enclosure mechanisms 7 on both sides form openings 8. The enclosure mechanisms 7 are set to limit the number of pets eating at the same time. The main control unit obtains the eating data of a single pet that has entered the enclosure mechanism 7 based on weight data and image data, thus avoiding inaccurate statistics caused by multiple pets eating at the same time.

[0074] The weight sensor is mounted on the base 2. The feeding data includes the weight of the food consumed. The method for obtaining the weight of the food using the weight sensor is as follows:

[0075] The main control unit determines whether there is a pet in the target area using a proximity sensor. If there is no pet, the main control unit controls the weight sensor to enter standby mode. If there is a pet, the main control unit uses a weighing sensor to measure the weight of the feeding trough 6 when the pet enters the target area and when the pet leaves the target area. The weight of the feeding trough 6 is the difference between its weight and the weight of the feeding trough 6. After the pet leaves the target area, the weight of the feeding trough 6 is measured again after it stabilizes. Stable weight of the feeding trough 6 means that its weight does not change within a certain period of time. In a preferred embodiment, the target area is the internal area of ​​the enclosure mechanism 7.

[0076] Feeding data includes the number of feedings, which is obtained in the following way:

[0077] The main control unit uses a proximity sensor to determine if a pet is present in the target area. If no pet is present, the main control unit activates the image sensor 5 and the weight sensor to enter standby mode. If a pet is present, the main control unit uses a weighing sensor to check if the weight data in the feeding trough 6 has changed. If the weight data remains unchanged, the feeding count is not recorded; if the weight data changes, the feeding count is recorded. By tracking the pet's feeding frequency, the system helps users understand the pet's eating habits and monitor its health.

[0078] Feeding data includes feeding duration, which is obtained in the following ways:

[0079] The main control unit uses a proximity sensor to determine if there is a pet in the target area; if there is no pet, the main control unit controls the image sensor 5 and the weight sensor to enter standby mode; if there is a pet, the main control unit uses a weighing sensor to determine if the weight data of the feeding trough 6 has changed.

[0080] If the weight data does not change, the current time is not included in the feeding time, and the image sensor 5 remains in standby mode; if the weight data changes, the main control unit uses the image sensor 5 to identify the pet's head posture.

[0081] If the pet's head is lowered, the current time is counted towards the feeding time; if the pet's head is not lowered, the current time is not counted towards the feeding time.

[0082] Alternatively, the duration of the meal can also be obtained in the following ways:

[0083] The main control unit uses a proximity sensor to determine if there is a pet in the target area; if there is no pet, the main control unit controls the image sensor 5 and the weight sensor to enter standby mode; if there is a pet, the main control unit uses a weighing sensor to determine if the weight data of the feeding trough 6 has changed.

[0084] If the weight data does not change, the current time is not included in the feeding time, and the image sensor 5 remains in standby mode. If the weight data changes, the period of time during which the weight data changes is the feeding time. When the weight data changes, the time point at which the change begins is recorded as the starting point. When the weight data stops changing, it is determined that the pet has finished eating, and this point is recorded as the time point at which the pet finishes eating. The difference between the two time points is the feeding time.

[0085] Since a pet's eating is not continuous, it can be a process of "eating, resting, eating" or "eating, resting, leaving". Therefore, the duration of a pet's eating is uncertain from the time it enters the target area to the time it leaves the target area.

[0086] In this invention, the weight sensor and the image sensor 5 cooperate to improve the accuracy of feeding time statistics and eliminate some cases that should not be included in the feeding time. This device solves the above-mentioned defects by combining the weight change of the feeding trough and the pet's head posture.

[0087] On the one hand, determining the pet's feeding time solely based on the duration of weight changes in the feeding trough 6 may be inaccurate. For example, if the pet is merely fiddling with the trough instead of eating, the weight change in the trough 6 should not be included in the feeding time. Similarly, determining the feeding time solely based on the duration the pet's head is lowered will also fail to provide an accurate estimate. For instance, if the pet is simply playing with its head down, the feeding time should not be counted, as the weight of the feeding trough 6 has not changed and the pet is not eating. Therefore, this invention uses a combination of weight changes in the feeding trough 6 and the image sensor 5 recognizing the pet's head posture as lowered to include the feeding time.

[0088] On the other hand, since the image sensor 5 consumes a lot of power when it is working, the feeding time statistics method disclosed in this invention can greatly reduce the working frequency of the image sensor 5 and improve the overall battery life. That is, if the pet is in the target area and the weight data has not changed, the current time is not included in the feeding time, and the image sensor 5 is kept in standby mode.

[0089] The food in the feeding trough 6 is fed in any of the following ways;

[0090] Method 1: The main control unit collects the weight data of the feeding trough 6 through a weighing sensor. When the weight data is lower than the threshold, the main control unit controls the feeding of food until the weight of the feeding trough 6 reaches the maximum set value.

[0091] Method 2: Set a single feeding amount for each pet. The main control unit feeds the pets identified based on image data in a single feeding. The single feeding amount is the single feeding amount minus the remaining food in the feeding trough 6.

[0092] The first feeding method can reduce food waste, while the second feeding method can control the amount of food each pet eats, allowing for flexible selection based on needs.

[0093] In a preferred embodiment, the automatic feeding device further includes an image server 9 and a business logic server 10. The image server 9 is used to store image data, and the business logic server 10 is used to display the pet's feeding data to the user. The specific method is as follows:

[0094] The image sensor 5 uploads image data to the image server 9, generates an image ID corresponding to the image data, and transmits the image ID back to the main control unit. The main control unit uploads the feeding data and image ID to the business logic server 10. The image ID can be a fragment of the image data or a specific marker, reducing the data load on the main control unit and facilitating the association between the image data and weight data. The logic server retrieves the image data from the image server 9 based on the image ID, obtains the pet ID corresponding to the image data through a pet identification network, and stores the association between the pet ID and the feeding data through the main control unit. The pet ID can be the pet's name. Users access the business logic server 10 through terminal devices to obtain the pet's feeding data. The image server 9 and business logic server 10 are deployed in the cloud, facilitating data management and reducing the hardware cost of the automatic feeding device itself.

[0095] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A pet identification based feeding system, characterized in that, The application relates to a feeding device for pets. The feeding device comprises a master control unit, an image sensor (5), a feeding trough (6) and a weight sensor. The weight sensor collects weight data of the feeding trough (6). The image sensor (5) collects image data of the pet eating. The master control unit obtains feeding data of the pet based on the image data according to the weight data. Two sides of the feeding trough (6) are provided with fence mechanisms (7), and the fence mechanisms (7) on the two sides are formed with openings (8). The master control unit obtains feeding data of a single pet entering the fence mechanism (7) based on the image data according to the weight data. The feeding data comprises feeding duration. The feeding duration is obtained by the following method: The master control unit judges whether there is a pet in a target area through a proximity sensor; if not, the master control unit controls the image sensor (5) and the weight sensor to enter a standby state; if yes, the master control unit judges whether the weight data of the feeding trough (6) changes through the weight sensor. If the weight data does not change, the current time is not counted into the feeding duration, and the image sensor (5) remains in the standby state; if the weight data changes, the master control unit identifies the head posture of the pet through the image sensor (5). If the head posture of the pet is low, the current time is counted into the feeding duration; if the head posture of the pet is not low, the current time is not counted into the feeding duration. When the weight data changes, the time point when the change starts is recorded as a time starting point; when it is detected that the weight data no longer changes, it is judged that the pet has finished eating, and the time point when the pet finishes eating is recorded; the difference between the two time points is the feeding duration. The feeding data comprises feeding weight. The feeding weight is obtained by the following method: The master control unit judges whether there is a pet in a target area through a proximity sensor; if not, the master control unit controls the weight sensor to enter a standby state; if yes, the master control unit measures the weight of the feeding trough (6) when the pet enters the target area and when the pet leaves the target area through the weight sensor, and the feeding weight is the difference between the weights of the feeding trough (6). The food in the feeding trough (6) is fed by any one of the following methods. Method one: the master control unit collects the weight data of the feeding trough (6) through the weight sensor, and controls the food to be fed when the weight data is lower than a threshold value until the weight of the feeding trough (6) reaches a maximum set value. Method two: a single feeding amount is set for each pet, and the master control unit feeds each pet once based on the image data, and the feeding amount is the single feeding amount minus the remaining food in the feeding trough (6). The weight data is collected when the feeding trough (6) is in a stable state, and the stable state is that the weight of the feeding trough (6) does not change within a set time.

2. The pet identification based feeding system of claim 1, wherein, The feeding data comprises feeding times. The feeding times are obtained by the following method: The master control unit determines whether there is a pet in the target area through the proximity sensor; if there is no pet, the master control unit controls the image sensor (5) and the weight sensor to enter a standby state; if there is a pet, the master control unit determines whether the weight data of the feeding trough (6) changes through the weight sensor; If the weight data does not change, the feeding times are not counted; if the weight data changes, the feeding times are counted.

3. The pet identification based feeding system of claim 1, wherein, The image data is analyzed through the pet identity recognition network to obtain the corresponding pet ID, and the master control unit associates the pet ID with the feeding data.

4. The pet identification based feeding system of claim 3, wherein, The pet recognition-based feeding system further comprises a business logic server (10); The business logic server (10) is used to store the feeding data and pet ID of the pet and display them through connection with the user terminal device; The pet recognition-based feeding system further comprises an image server (9); The business logic server (10) obtains the pet ID and feeding data in the following manner: The image sensor (5) uploads the image data to the image server (9) to generate an image ID corresponding to the image data, and the master control unit uploads the feeding data and the image ID to the business logic server (10); The business logic server calls the image data from the image server (9) according to the image ID, obtains the pet ID corresponding to the image data through the pet identity recognition network, and associates the pet ID with the feeding data.

5. A pet feeding method, characterized by, The pet recognition-based feeding system of any one of claims 1 to 4 is used for feeding.

Images