Pet excrement monitoring equipment based on intelligent sensor

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to pet excrement monitoring equipment based on an intelligent sensor, which comprises an excrement detector, a shape analyzer, a result judging device and an early warning device, and is characterized in that the excrement detector is used for collecting an excrement image of a pet toilet, and the shape analyzer is used for identifying the edge contour of the excrement image; the result determiner is used for analyzing the faeces image according to the shape category labels set by the faeces image and determining whether feature elimination needs to be carried out based on dominant anomaly characterization values calculated based on surface gloss features, and the result determiner is used for determining whether feature elimination needs to be carried out based on the dominant anomaly characterization values calculated based on the surface gloss features. According to the pet excrement monitoring device, the interference feature is eliminated, whether the excrement image is abnormal or not is judged based on the disease characterization feature after the interference feature is eliminated, the accuracy of the pet excrement monitoring device is improved, the early warning device determines whether the excrement image is abnormal or not based on the disease characterization feature extracted by the result judging device, and the efficiency of the pet excrement monitoring device is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent monitoring, and particularly to a pet feces monitoring device based on intelligent sensors. Background Art

[0002] As pet health monitors have become a new technology in the field of pet care, they have become increasingly popular among pet owners in recent years. These devices typically use non-invasive sensor technology to monitor the vital signs of pets and transmit data wirelessly to a smartphone application.

[0003] Chinese Patent Publication No.: CN117694267A, discloses a pet status monitoring method, device and equipment, including: obtaining monitoring information of a pet, where the monitoring information includes monitoring video and / or usage data of sensors in pet utensils; performing pet behavior recognition on the monitoring information, and obtaining corresponding behavior status information based on the recognized pet behavior, and summarizing to obtain pet status information according to the pet behavior and the corresponding behavior status information; converting the pet status information into a text form, and forming a user reminder according to the pet status information in text form.

[0004] However, the following problems still exist in the prior art:

[0005] In the prior art, the problem that the accuracy of pet feces monitoring is low due to the lack of consideration of setting monitoring devices in pet toilets to monitor pet feces is not considered;

[0006] The problem that the efficiency of pet monitoring is low due to the lack of consideration of the influence of the freshness of pet feces on the determination of pet diseases is not considered. Summary of the Invention

[0007] To solve the above problems, the present invention provides a pet feces monitoring device based on intelligent sensors, which overcomes the problems in the prior art that the accuracy of pet feces monitoring is low due to the lack of consideration of setting monitoring devices in pet toilets to monitor pet feces, and the efficiency of pet monitoring is low due to the lack of consideration of the influence of the freshness of pet feces on the determination of pet diseases.

[0008] To achieve the above object, the present invention provides a pet feces monitoring device based on intelligent sensors, including:

[0009] A feces detector, including an image acquisition unit disposed in the pet toilet for acquiring images of feces in the pet toilet;

[0010] A shape analyzer, which is connected to the feces detector, for identifying the edge contour of the feces image and setting a shape category label for the feces image according to the curvature of each contour segment of the edge contour;

[0011] A result determiner, which is connected to the shape analyzer and the feces detector, and is used to extract features from the feces image according to the shape category label set for the feces image, including,

[0012] Extract the surface gloss feature in the current feces image, calculate the dominant anomaly characterization value according to the surface gloss feature, determine whether feature elimination is required based on the dominant anomaly characterization value, and extract the disease characterization features in the feces image with interference features eliminated or without interference features eliminated;

[0013] Or, extract the disease characterization features in the feces image;

[0014] An early warning device, which is used to determine whether the feces image is abnormal based on the disease characterization features extracted by the result determiner, so as to issue an early warning signal;

[0015] Among them, the feces image includes a surface gloss feature and a color feature. The surface gloss feature includes the brightness feature and the color saturation feature of the feces image; the feature elimination includes identifying the cross-section specific area and eliminating interference features according to the cross-section specific area; the color feature includes the average chromaticity of the feces and the color difference of the feces.

[0016] Further, the shape analyzer is used to set a shape category label for the feces image based on the curvature of each contour segment of the edge contour. Among them,

[0017] Used to calculate the average value of the curvature of each contour segment of the edge contour of the feces image;

[0018] If the average value of the curvature of each contour segment of the edge contour is greater than or equal to the predetermined curvature threshold, it is determined that the shape category of the feces image is the first category label;

[0019] If the average value of the curvature of each contour segment of the edge contour is less than the predetermined curvature threshold, it is determined that the shape category of the feces image is the second category label.

[0020] Further, the result determiner analyzes the feces image based on the shape category label set for the feces image, including,

[0021] If it is determined that the shape category is the first category label, extract the surface gloss feature in the current feces image, calculate the dominant anomaly characterization value according to the surface gloss feature, determine whether feature elimination is required based on the dominant anomaly characterization value, and extract the disease characterization features in the feces image with interference features eliminated or without interference features eliminated;

[0022] If it is determined that the shape category is the second category label, extract the disease characterization features in the feces image.

[0023] Further, the result determiner is used to calculate a dominant anomaly characterization value based on the surface gloss characteristics, including:

[0024] Calculating the ratio of the brightness of the fecal image to the brightness threshold value and determining it as the first dominant anomaly data feature;

[0025] Calculating the ratio of the color saturation of the fecal image to the saturation threshold value and determining it as the second dominant anomaly data feature;

[0026] Calculating the sum of the first dominant anomaly data feature and the second dominant anomaly data feature and determining it as the dominant anomaly characterization value.

[0027] Further, the result determiner is used to determine whether feature elimination is required based on the dominant anomaly characterization value, including:

[0028] If the dominant anomaly characterization value is less than or equal to the predetermined dominant anomaly characterization value, it is determined that feature elimination is required.

[0029] Further, identifying the cross-section specific area and eliminating interference features based on the cross-section specific area, including:

[0030] Identifying the fecal cross-section contour in the fecal image and determining the fecal cross-section contour as the cross-section specific area;

[0031] Using the non-cross-section feature area as the interference feature to be eliminated.

[0032] Further, the result determiner is also used to identify the disease symptom characterization features, including:

[0033] Identifying the fecal color difference;

[0034] Identifying the difference between the average chromaticity of the feces and the preset standard chromaticity;

[0035] Identifying the variance of the curvature of each contour segment of the edge contour in the fecal image.

[0036] Further, the result determiner is used to determine the feature anomaly characterization according to the disease symptom characterization features, including:

[0037] Calculating the difference between the average chromaticity of the feces and the preset standard chromaticity and determining it as the first disease symptom characterization feature;

[0038] Calculating the variance of the curvature of each contour segment of the edge contour in the fecal image and determining it as the second disease symptom characterization feature,

[0039] Calculating the sum of the first disease symptom characterization feature and the second disease symptom characterization feature and determining it as the feature anomaly characterization parameter.

[0040] Further, the warning device determines whether the fecal image is abnormal based on the disease symptom characteristics extracted by the result determiner, including

[0041] If the characteristic abnormal representation parameter is greater than or equal to the preset characteristic abnormal representation threshold, it is determined that the fecal image is abnormal.

[0042] Further, the warning device is used to send a warning signal to the mobile terminal.

[0043] Compared with the prior art, the present invention provides a pet feces monitoring device based on an intelligent sensor, including a feces detector, a shape analyzer, a result determiner and a warning device. The feces detector is used to collect the fecal image of the pet toilet. The shape analyzer is used to identify the edge contour of the fecal image and set a shape category label for the fecal image according to the curvature of each contour segment of the edge contour. The result determiner is used to analyze the fecal image based on the shape category label set for the fecal image, determine whether feature elimination is required based on the dominant abnormal representation value calculated based on the surface gloss feature, and determine whether there is an abnormality in the fecal image based on the disease symptom characteristics after excluding interference features, improving the accuracy of the pet feces monitoring device. Moreover, the warning device determines whether the fecal image is abnormal based on the disease symptom characteristics extracted by the result determiner to send a warning signal, further improving the efficiency of the pet feces monitoring device.

[0044] In particular, the present invention identifies the edge contour of the fecal image through the shape analyzer, and can distinguish the shape category label through the curvature of each contour segment of the edge contour, including the first category label of the formed fecal shape and the second category label of the unformed fecal shape. Since the fecal data of different categories are different, it provides data support for adaptively adopting different image analysis methods subsequently, improving the reliability and accuracy of analyzing pet fecal diseases.

[0045] In particular, the present invention can analyze the fecal image through the result determiner to respectively process whether there is an abnormality in the fecal image in the first category label and the second category label, and determine whether to eliminate interference features through the surface glossiness feature. Since the freshness of the feces decreases, the surface moisture evaporates, resulting in a decrease in the surface glossiness, making the feces more homogeneous and difficult to identify disease symptom features. Therefore, it is considered to eliminate interference features, and considering that the fecal cross-section contour has strong data representativeness, the disease symptom features are identified based on this, thereby improving the accuracy and efficiency of feces monitoring. Description of the Drawings

[0046] Figure 1 It is a structural block diagram of the pet feces monitoring device based on an intelligent sensor according to an embodiment of the present invention;

[0047] Figure 2Logic decision diagram for setting shape category tags in the embodiments of the present invention;

[0048] Figure 3 Logic decision diagram for analyzing the shape category tags of fecal images in the embodiments of the present invention;

[0049] Figure 4 Logic decision diagram for calculating the dominant anomaly characterization value based on the surface gloss feature in the embodiments of the present invention. Detailed implementation manners

[0050] In order to make the objectives and advantages of the present invention clearer and more understandable, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0051] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0052] It should be noted that in the description of the present invention, unless otherwise clearly defined and limited, the term "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific situations.

[0053] Please refer to Figure 1 As shown, it is a structural block diagram of a pet feces monitoring device based on an intelligent sensor in the embodiments of the present invention. The present invention provides a pet feces monitoring device based on an intelligent sensor, including:

[0054] A feces detector, including an image acquisition unit arranged in the pet toilet for acquiring fecal images in the pet toilet;

[0055] A shape analyzer, which is connected to the feces detector and is used to identify the edge contour of the fecal image and set a shape category tag for the fecal image according to the curvature of each contour segment of the edge contour;

[0056] A result determiner, which is connected to the shape analyzer and the feces detector and is used to extract features from the fecal image according to the shape category tag set for the fecal image, including,

[0057] Extract the surface gloss features in the current fecal image, calculate the dominant abnormal characterization value based on the surface gloss features, determine whether feature elimination is required based on the dominant abnormal characterization value, and extract the disease characterization features in the fecal image with interference features eliminated or without interference features eliminated;

[0058] Or, extract the disease characterization features in the fecal image;

[0059] An early warning device, which is used to determine whether the fecal image is abnormal based on the disease characterization features extracted by the result determination device, so as to emit an early warning signal;

[0060] Among them, the fecal image includes surface gloss features and color features. The surface gloss features include the brightness feature and color saturation feature of the fecal image; the feature elimination includes identifying the cross-section specific area and eliminating interference features based on the cross-section specific area; the color features include the average chromaticity and color difference.

[0061] Specifically, the specific structure of the fecal detector is not limited. It can use a camera, as long as it can capture images, which will not be elaborated here.

[0062] Specifically, the specific structures of the shape analyzer, result determination device, and early warning device are not limited. They can all be composed of logic components. The logic components include field programmable components, computers, or microprocessors in computers.

[0063] It can be understood that the shape category label includes a first label and a second label. Preferably, the first label represents the feces with a formed fecal shape, and the second label represents the feces with an unformed fecal shape.

[0064] It can be understood that the standard color distribution is the average value of the yellow-brown color of the feces collected within the historical period as the standard color distribution. The fecal color difference mainly refers to the difference between the fecal color and the normal color. For example, the color difference between the fecal color and the normal standard fecal color.

[0065] Specifically, the pet fecal monitoring device of the present invention includes a fecal detector, a shape analyzer, a result determination device, and an early warning device. The fecal detector can collect pet toilet fecal images. The shape analyzer can identify the edge contour of the fecal image. The shape category label of the fecal image can be analyzed through the curvature of each contour segment of the edge contour. The result determination device can analyze the fecal image based on the shape category label set for the fecal image, and determine whether there is an abnormality in the fecal image based on the disease characterization features, improving the accuracy of the pet fecal monitoring device. Moreover, when the early warning device determines that the fecal image is abnormal based on the disease characterization features extracted by the result determination device, it can emit an early warning signal, further improving the efficiency of the pet fecal monitoring device.

[0066] Please refer to Figure 2 as shown, which is the logical decision diagram for setting the shape category label in the embodiment of the present invention. The shape analyzer of the present invention is used to set the shape category label for the fecal image based on the curvature of each contour segment of the edge contour, where

[0067] is used to calculate the average value of the curvatures of each contour segment of the edge contour of the fecal image;

[0068] If the average value of the curvatures of each contour segment of the edge contour is greater than or equal to the predetermined curvature threshold, it is determined that the shape category of the fecal image is the first category label;

[0069] If the average value of the curvatures of each contour segment of the edge contour is less than the predetermined curvature threshold, it is determined that the shape category of the fecal image is the second category label.

[0070] It can be understood that the predetermined curvature threshold is 1.15 to 1.35 times, preferably 1.23 times, the average value of the curvatures of each contour segment within the historical period.

[0071] It can be understood that the smaller the average value of the curvatures of each contour segment of the edge contour, the closer the fecal shape is to the unformed state; conversely, the larger the average value of the curvatures of each contour segment of the fecal edge contour, the closer the fecal shape is to the formed state.

[0072] Specifically, there is no limitation on the specific method for identifying the edge contour. The edge contour can be identified by using an image segmentation algorithm. Of course, other methods can also be used, which will not be elaborated here.

[0073] Specifically, the present invention identifies the edge contour of the fecal image through the shape analyzer, and can distinguish the shape category label through the curvatures of each contour segment of the edge contour, including the first category label for the formed fecal shape and the second category label for the unformed fecal shape. Since the fecal data of different categories show different performances, therefore, it provides data support for adaptively adopting different image analysis methods subsequently, and improves the reliability and accuracy of the analysis for pet fecal diseases.

[0074] Please refer to Figure 3 as shown, which is the logical decision diagram for analyzing the shape category label of the fecal image in the embodiment of the present invention. The result determiner of the present invention analyzes the fecal image based on the shape category label set for the fecal image, including,

[0075] If it is determined that the shape category is the first category label, the surface gloss feature in the current fecal image is extracted, the dominant abnormal characterization value is calculated based on the surface gloss feature, and it is determined whether feature elimination is required based on the dominant abnormal characterization value, and the disease characterization feature in the fecal image with the interference feature eliminated or not eliminated is extracted;

[0076] If it is determined that the shape category is the second category label, the disease characterization features in the fecal image are extracted.

[0077] Please refer to Figure 4 As shown, it is a logical decision diagram for calculating the dominant abnormal characterization value based on the surface gloss feature in an embodiment of the present invention. The result determiner is used to calculate the dominant abnormal characterization value based on the surface gloss feature, including,

[0078] Determining the ratio of the brightness of the fecal image to the brightness threshold as the first dominant abnormal data feature;

[0079] Determining the ratio of the color saturation of the fecal image to the saturation threshold as the second dominant abnormal data feature;

[0080] Determining the sum of the first dominant abnormal data feature and the second dominant abnormal data feature as the dominant abnormal characterization value.

[0081] It can be understood that the brightness threshold is 1.15 times the average brightness of the fecal images in the historical data, and the saturation threshold is 1.2 times the average saturation of the fecal images in the historical data.

[0082] Specifically, the result determiner is used to determine whether feature elimination is required based on the dominant abnormal characterization value, including,

[0083] If the dominant abnormal characterization value is less than or equal to the predetermined dominant abnormal characterization value, it is determined that feature elimination is required.

[0084] It can be understood that the dominant abnormal characterization value is a judgment criterion that combines the first dominant abnormal data feature and the second dominant abnormal data feature. In practice, the brightness and color saturation of the fecal images in the historical data are values obtained based on fresh feces as a reference. When the fecal image of the contour segment is abnormal, the smaller the brightness of the fecal image, the smaller the ratio of the brightness of the fecal image to the brightness threshold, that is, the smaller the first dominant abnormal data feature. Similarly, the smaller the color saturation of the fecal image, the smaller the second dominant abnormal data feature, indicating that there is an abnormal area in the fecal image, that is, feature elimination is required.

[0085] Specifically, identifying the cross-section specific area and eliminating interference features based on the cross-section specific area, including,

[0086] Identifying the fecal cross-section contour in the fecal image and determining the fecal cross-section contour as the cross-section specific area;

[0087] Using the non-cross-section feature area as an interference feature to be eliminated.

[0088] Specifically, an image processing algorithm or model capable of recognizing the cross-sectional profile can be pre-trained and imported into the logic component to implement the corresponding function, which will not be elaborated here.

[0089] Specifically, the result determiner is further configured to recognize the disease symptom characteristics, including

[0090] being configured to recognize the color difference of feces;

[0091] being configured to recognize the difference between the average chromaticity of feces and the preset standard chromaticity;

[0092] being configured to recognize the variance of the curvature of each contour segment of the edge contour in the feces image.

[0093] It can be understood that the pet disease symptoms include but are not limited to gastroenteritis, a digestive system disease.

[0094] It can be understood that the preset standard chromaticity is the average value of the chromaticity data in the historical data monitored by the pet feces monitoring device in the previous three months within the historical period.

[0095] Specifically, the result determiner is configured to determine the characteristic abnormal representation according to the disease symptom characteristics, including

[0096] being configured to calculate the difference between the average chromaticity of feces and the preset standard chromaticity and determine it as the first disease symptom characteristic;

[0097] being configured to calculate the variance of the curvature of each contour segment of the edge contour in the feces image and determine it as the second disease symptom characteristic,

[0098] being configured to calculate the sum of the first disease symptom characteristic and the second disease symptom characteristic and determine it as the characteristic abnormal representation parameter.

[0099] It can be understood that the preset standard chromaticity difference is 1.15 times the average value of the chromaticity differences in the historical data monitored by the pet feces monitoring device in the previous three months within the historical period.

[0100] Specifically, the warning device determines whether the feces image is abnormal based on the disease symptom characteristics extracted by the result determiner, including

[0101] If the characteristic abnormal representation parameter is less than the preset characteristic abnormal representation threshold, it is determined that the feces image is not abnormal;

[0102] If the characteristic abnormal representation parameter is greater than or equal to the preset characteristic abnormal representation threshold, it is determined that the feces image is abnormal.

[0103] It can be understood that the preset characteristic abnormal representation threshold is 1.05 times the average value of the characteristic abnormal representation parameters in the historical monitoring data.

[0104] Specifically, the pet feces monitoring device based on intelligent sensors provided by the present invention further includes an early warning device for sending an early warning signal to a mobile terminal.

[0105] So far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A pet feces monitoring device based on intelligent sensors, characterized in that: include: A feces detector, comprising an image acquisition unit disposed in the pet toilet for acquiring images of feces in the pet toilet; a shape analyzer connected to the stool detector and used to identify the edge contour of the stool image and set a shape category label for the stool image according to the curvature of each contour segment of the edge contour; A result determiner is connected to the shape analyzer and the feces detector and is used to extract features of the feces image according to the shape category label set for the feces image, including: Extracting surface gloss features in the current stool image, calculating dominant abnormality characterization values ​​based on the surface gloss features, determining whether feature removal is required based on the dominant abnormality characterization values, and extracting disease characterization features in the stool image with or without interference features removed; or, extracting disease-representing features from stool images; An early warning device, which is used to determine whether the stool image is abnormal based on the disease characterization features extracted by the result determiner, so as to issue an early warning signal; Among them, the feces image includes surface gloss features and color features, and the surface gloss features include brightness features and color saturation features of the feces image; the feature elimination includes identifying cross-section specific areas and eliminating interference features based on cross-section specific areas; the color features include average feces chromaticity and feces color difference.

2. The pet feces monitoring device based on smart sensor according to claim 1 is characterized in that: The shape analyzer is used to set a shape category label for the stool image based on the curvature of each contour segment of the edge contour, wherein: Used to calculate the average value of the curvature of each contour segment of the edge contour of the feces image; If the average value of the curvature of each contour segment of the edge contour is greater than or equal to a predetermined curvature threshold, then determining that the shape category of the feces image is a first category label; If the average value of the curvatures of the contour segments of the edge contour is less than a predetermined curvature threshold, the shape category of the feces image is determined to be a second category label.

3. The pet feces monitoring device based on smart sensor according to claim 1 is characterized in that: The result determiner analyzes the feces image based on the shape category label set for the feces image, including: If the shape category is determined to be a first category label, then extracting the surface gloss feature in the current stool image, calculating the dominant abnormality characterization value based on the surface gloss feature, determining whether feature removal is required based on the dominant abnormality characterization value, and extracting the disease characterization feature in the stool image with or without removing the interference feature; If the shape category is determined to be the second category label, then the disease characterizing features in the stool image are extracted.

4. The pet feces monitoring device based on smart sensor according to claim 3 is characterized in that: The result determiner is used to calculate the dominant abnormality characterization value based on the surface gloss feature, including: The ratio of the brightness of the stool image to the brightness threshold is used to calculate and determine as the first dominant abnormal data feature; The ratio of the color saturation of the stool image to the saturation threshold is used to calculate the second dominant abnormal data feature; The sum of the first dominant abnormal data feature and the second dominant abnormal data feature is calculated to be determined as the dominant abnormal characterization value.

5. The pet feces monitoring device based on smart sensor according to claim 1, characterized in that: The result determiner is used to determine whether feature removal is required based on the dominant abnormality characterization value, including: If the dominant abnormality characterization value is less than or equal to the predetermined dominant abnormality characterization value, it is determined that feature elimination is required.

6. The pet feces monitoring device based on smart sensor according to claim 1, characterized in that: Identify cross-section-specific regions and remove interfering features based on cross-section-specific regions, including: for identifying a feces cross-sectional contour in the feces image, and determining the feces cross-sectional contour as a cross-sectional specific region; Used to remove non-section feature areas as interference features.

7. The pet feces monitoring device based on smart sensor according to claim 1, characterized in that: The result determiner is also used to identify symptoms, including: Used to identify stool color differences; Used to identify the difference between the average color of feces and the preset standard color; Used to identify the variance of the curvature of each contour segment of the edge contour in the feces image.

8. The pet feces monitoring device based on smart sensor according to claim 1, characterized in that: The result determiner is used to determine the abnormal characteristic according to the symptom characteristic feature, include, The difference between the average color of stool and the preset standard color is used to determine as the first disease characterization feature; The variance of the curvature of each contour segment of the edge contour in the feces image is used to calculate the second disease characterization feature. The sum of the first disease characterization feature and the second disease characterization feature is calculated to be determined as the characteristic abnormality characterization parameter.

9. The pet feces monitoring device based on smart sensor according to claim 1, characterized in that: The early warning device determines whether the stool image is abnormal based on the disease characterization features extracted by the result determiner, including: If the characteristic abnormality characterization parameter is greater than or equal to the preset characteristic abnormality characterization threshold, the stool image is determined to be abnormal.

10. The pet feces monitoring device based on smart sensor according to claim 9, characterized in that: The warning device is used to send a warning signal to the mobile terminal.

Images