Intelligent assessment methods, devices, equipment and storage media for infant and toddler meal quality

Abstract

This invention relates to the field of intelligent childcare, solving the problem of existing technologies' inability to effectively assess the quality of infants' and toddlers' eating behavior. It provides an intelligent assessment method, device, equipment, and storage medium for infants' and toddlers' eating quality. The method includes: acquiring a real-time video stream in an infant care scenario; decomposing the video stream into multiple frames; pre-setting a target detection model; inputting the multiple frames into the target detection model; identifying the position information of a person's hands and the position information of the infant's mouth in the multiple frames; and based on the position information of the hands and the mouth, identifying whether the infant is eating; when the infant is identified as eating, analyzing the infant's eating behavior according to preset behavior analysis rules to assess the quality of the infant's eating. This invention helps users guide infants' and toddlers during meals, improves the standardization of infants' and toddlers' meals, and also enhances the care experience for parents.

Description

Technical Field

[0001] This invention relates to the field of intelligent care, and in particular to an intelligent assessment method, device, equipment, and storage medium for infants' and toddlers' meal quality. Background Technology

[0002] With the development and popularization of various smart terminals, the application of smart care devices is becoming more and more widespread, gradually becoming a part of people's lives.

[0003] In existing technologies, high chairs for infants and toddlers mainly provide physical convenience for eating. However, the analysis of infants' and toddlers' eating behavior mainly comes from the parents' manual supervision. Some smart high chairs for infants and toddlers are equipped with a smart camera to record the infants' and toddlers' eating habits. However, the analysis of the infants' and toddlers' eating behavior is only simple and rough, and the analysis results are extremely incomplete. It is impossible to guide parents to adjust and correct the infants' and toddlers' eating behavior based on the analysis results. As a result, the actual care experience obtained by parents is poor.

[0004] Therefore, how to effectively assess the quality of infants' and toddlers' eating behavior is an urgent problem to be solved. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide an intelligent assessment method, device, equipment, and storage medium for infants' and toddlers' meal quality, in order to solve the problem that the prior art cannot effectively assess the quality of infants' and toddlers' meal behavior.

[0006] In a first aspect, embodiments of the present invention provide an intelligent assessment method for the quality of infant and toddler meals, characterized in that the method includes:

[0007] S1: Acquire a real-time video stream in an infant care scenario and decompose the video stream into multiple frames of images;

[0008] S2: Pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hand and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hand and the position information of the infant's mouth.

[0009] S3: When an infant or toddler is identified eating, the system analyzes the infant or toddler's eating behavior based on preset behavior analysis rules and evaluates the quality of the infant or toddler's meal.

[0010] Preferably, S3 includes:

[0011] S31: When recognizing infants and young children eating, extract the target images corresponding to each frame of the infants and young children eating;

[0012] S32: Based on the hand position information of the person and the mouth position information of the infant, the number of times the infant eats is calculated;

[0013] S33: Based on the preset meal quality assessment rules and combined with the number of times the infants and young children eat, a comprehensive analysis is performed on each of the target images to assess the meal quality of the infants and young children.

[0014] Preferably, S33 includes:

[0015] S331: Perform an autonomy analysis on the infant's eating behavior in the target image, and obtain an autonomy score by combining the autonomy analysis results with the number of times the infant eats;

[0016] S332: Perform a pleasure analysis on the infants' eating behavior in the target image, and obtain a pleasure score by combining the pleasure analysis results with the number of times the infants eat;

[0017] S333: Perform compactness analysis on the infant and toddler eating behavior in the target image, and obtain a compactness score by combining the compactness analysis results with the number of times the infant and toddler eats.

[0018] S334: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score. Based on the comprehensive score, the evaluation result is given.

[0019] Preferably, S331 includes:

[0020] S3311: A first classification model is pre-set, and the human hand is classified into infant hand and non-infant hand using the first classification model;

[0021] S3312: Obtain the number of times the infant's hands appear, and count the number of times the infant eats independently based on the number of times the infant's hands appear;

[0022] S3313: The autonomy score is calculated by combining the number of times the infant eats and the number of times the infant eats independently.

[0023] Preferably, S332 includes:

[0024] S3321: Perform face detection on the target image to identify the facial information of the infant in the target image;

[0025] S3322: A second classification model is pre-set, and the infant's facial information is input into the second classification model to identify the infant's facial expression information;

[0026] S3323: Based on the aforementioned facial expression information, count the number of times the infant smiled and cried;

[0027] S3324: Calculate the pleasure score by combining the number of times the infant smiled, cried, and ate.

[0028] Preferably, S333 includes:

[0029] S3331: Obtain the timing of infant feedings;

[0030] S3332: Based on the stated time points, calculate the average interval time and time series;

[0031] S3333: Based on the average interval time, time series and the number of times the infants and young children eat, calculate the compactness score.

[0032] Preferably, S334 includes:

[0033] S3341: Obtain the age of the infant or toddler and pre-set the score threshold and age threshold;

[0034] S3342: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score;

[0035] S3343: Compare the overall score with the score threshold, and compare the infant's age with the age threshold to obtain the assessment result of the infant's meal quality.

[0036] Secondly, embodiments of the present invention also provide an intelligent assessment device for the quality of infant and toddler meals, characterized in that the device comprises:

[0037] The image acquisition module is used to acquire real-time video streams in infant and toddler care scenarios and decompose the video streams into multiple frames of images.

[0038] The behavior determination module is used to pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hands and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hands and the position information of the infant's mouth.

[0039] The behavior assessment module is used to analyze the eating behavior of infants and toddlers based on preset behavior analysis rules when they are identified eating, and to assess the quality of their meals.

[0040] Thirdly, embodiments of the present invention also provide an electronic device, including: at least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method of the first aspect described above.

[0041] Fourthly, embodiments of the present invention also provide a storage medium storing computer program instructions thereon, which, when executed by a processor, implement the method of the first aspect described above.

[0042] In summary, the beneficial effects of the present invention are as follows:

[0043] The intelligent assessment method, device, equipment, and storage medium for infant and toddler meal quality provided in this invention acquires real-time video streams in infant and toddler care scenarios and decomposes the video streams into multiple frames. A target detection model is pre-set, and the multiple frames are input into the model to identify the position information of a person's hands and the position information of the infant's mouth in the multiple frames. Based on the position information of the person's hands and the infant's mouth, it is determined whether the infant is eating. When the infant is identified as eating, the eating behavior is analyzed according to preset behavior analysis rules to assess the quality of the infant's meal. When infant and toddler eating occurs, the preset behavior analysis rules provide a relatively comprehensive analysis and assessment of the infant's eating behavior, yielding detailed assessment results. This helps users guide infants and toddlers during meals, improves the standardization of infants' meals, and enhances the care experience for parents. Attached Figure Description

[0044] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the embodiments of the present invention will be briefly introduced below. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort, and these are all within the protection scope of the present invention.

[0045] Figure 1 This is a flowchart illustrating the intelligent assessment method for infant and toddler meal quality in Embodiment 1 of the present invention;

[0046] Figure 2 This is a schematic diagram illustrating the determination of whether an infant is in a high chair in Embodiment 1 of the present invention;

[0047] Figure 3 This is a schematic diagram illustrating the determination of whether an infant or toddler is eating in Embodiment 1 of the present invention;

[0048] Figure 4 This is a schematic diagram of the process for assessing infants' and toddlers' eating behavior in Embodiment 1 of the present invention;

[0049] Figure 5 This is a schematic diagram of the weighted calculation of scores in Embodiment 1 of the present invention;

[0050] Figure 6 This is a schematic diagram of the process for determining dining autonomy scores in Embodiment 1 of the present invention;

[0051] Figure 7 This is a schematic diagram illustrating the determination of whether an infant is feeding independently in Embodiment 1 of the present invention;

[0052] Figure 8 This is a schematic diagram of the process for determining the dining pleasure score in Embodiment 1 of the present invention;

[0053] Figure 9 This is a schematic diagram of the process for determining the dining compactness score in Embodiment 1 of the present invention;

[0054] Figure 10 This is a flowchart illustrating the comprehensive evaluation results given in Embodiment 1 of the present invention;

[0055] Figure 11 This is a structural block diagram of the intelligent assessment device for infant and toddler meal quality in Embodiment 2 of the present invention;

[0056] Figure 12 This is a schematic diagram of the electronic device in Embodiment 3 of the present invention;

[0057] The numbers in the diagram are as follows:

[0058] 1—The smallest bounding rectangle of the infant's body; 2—The smallest bounding rectangle of the high chair; 3—The smallest bounding rectangle of the hand; 4—The smallest bounding rectangle of the infant's mouth; 5—The smallest bounding rectangle of the infant's hand. Detailed Implementation

[0059] The features and exemplary embodiments of various aspects of the present invention will now be described in detail. To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only configured to explain the present invention and are not configured to limit the present invention. For those skilled in the art, the present invention can be practiced without some of these specific details. The following description of the embodiments is merely intended to provide a better understanding of the present invention by illustrating examples of the invention.

[0060] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0061] Example 1

[0062] Please see Figure 1 Embodiment 1 of the present invention provides an intelligent assessment method for the quality of infants' and toddlers' meals, the method comprising:

[0063] S1: Acquire a real-time video stream in an infant care scenario and decompose the video stream into multiple frames of images;

[0064] Specifically, real-time video streams are acquired in infant and toddler care scenarios, wherein the video streams refer to color videos taken during the day and infrared video stream images taken at night, thereby achieving a 24-hour care experience for infants and toddlers, and the video streams are decomposed into multiple frames of images.

[0065] S2: Pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hand and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hand and the position information of the infant's mouth.

[0066] Specifically, the process begins by evaluating multiple frames of images derived from the decomposition of the real-time video stream. Images showing infants eating are then input into the next step for further evaluation of their eating behavior. Images where infants are not eating are discarded without further processing. Since the number of images from the real-time video stream decomposition may be large, performing quality analysis and evaluation on all of them would result in unnecessary resource waste and low efficiency. Discarding images where infants are not eating helps reduce unnecessary workflows and improves the efficiency of subsequent quality assessment. A large number of images from infant care scenarios are pre-collected, and the infant's body shape and high chair are labeled in these images. A deep learning algorithm is used to train the model, outputting a first detection model based on YOLOv5s that can identify the infant's body shape and high chair position. The multiple frames of images are then input into this first detection model. (See [link to relevant documentation]). Figure 2, the body position Pos1(x1, y1, w1, h1) of the infant and the position Pos3(x3, y3, w3, h3) of the dining chair are obtained, where x1 and x3 respectively represent the abscissa of the center point of the smallest circumscribed rectangle of the infant's body with label 1 and the abscissa of the center point of the smallest circumscribed rectangle of the dining chair with label 2; y1 and y3 respectively represent the ordinate of the center point of the smallest circumscribed rectangle of the infant's body and the ordinate of the center point of the smallest circumscribed rectangle of the dining chair; w1 and w3 respectively represent the width of the smallest circumscribed rectangle of the infant's body and the width of the smallest circumscribed rectangle of the dining chair; h1 and h3 respectively represent the height of the smallest circumscribed rectangle of the infant's body and the height of the smallest circumscribed rectangle of the dining chair. The above Pos1 and Pos3 are judged: if x1 > x3 and x1 < x3 + w3 and y1 > y3 are satisfied, it is considered that the infant is sitting on the dining chair; if x1 > x3 and x1 < x3 + w3 and y1 > y3 are not satisfied, it is considered that the infant is not sitting on the dining chair. Before judging whether the infant is dining, it is first judged whether the infant is sitting on the dining chair, which avoids the mis-start of the dining quality evaluation device when there is no infant sitting on the dining chair, effectively saves resources, and reduces unnecessary work consumption. When the infant is sitting on the dining chair, in the above-mentioned large number of images of the infant care scenario collected, the hands and the infant's mouth of the people in these images are marked, and trained using a deep learning algorithm to output a second detection model based on yolov5s that can identify the positions of the hands and the infant's mouth. Input the multi-frame images obtained by decomposing the real-time video stream into the second detection model, please refer to Figure 3 , the hand position is identified as Pos2(x2, y2, w2, h2) and the infant's mouth position is Pos4(x4, y4, w4, h4), where x2 and x4 respectively represent the abscissa of the center point of the smallest circumscribed rectangle of the hand with label 3 and the abscissa of the center point of the smallest circumscribed rectangle of the infant's mouth with label 4; y2 and y4 respectively represent the ordinate of the center point of the smallest circumscribed rectangle of the hand and the ordinate of the center point of the smallest circumscribed rectangle of the infant's mouth; w2 and w4 respectively represent the width of the hand and the width of the smallest circumscribed rectangle of the infant's mouth; h2 and h4 respectively represent the height of the smallest circumscribed rectangle of the hand and the height of the smallest circumscribed rectangle of the infant's mouth. At this time, Pos2 and Pos4 are further judged: for example, the horizontal distance threshold is set to 5 and the vertical distance threshold is set to 4. If the absolute value of the difference between x2 and x4 is less than 5 and the absolute value of the difference between y2 and y4 is less than 4, it is considered that the infant is dining on the dining chair; if the absolute value of the difference between x2 and x4 is not less than 5 and the absolute value of the difference between y2 and y4 is not less than 4, it is considered that the infant is on the dining chair but not dining. By further judging the positions of the hands and the infant's mouth, it is more accurate to judge whether the infant is dining, avoiding the situation that the infant is sitting on the dining chair but not dining, and improving the effectiveness of the subsequent evaluation of the infant's dining quality.

[0067] S3: When an infant or toddler is identified eating, the system evaluates the infant or toddler's eating behavior based on preset behavior analysis rules and outputs the evaluation results.

[0068] Specifically, when identifying infants and toddlers during mealtimes, the system comprehensively and effectively assesses their eating behavior based on pre-defined behavioral analysis rules, outputting evaluation results for user reference. This process, through the pre-defined behavioral analysis rules, provides a comprehensive and effective assessment of infants' and toddlers' eating behavior, assisting parents in correcting unhealthy eating habits. Furthermore, it helps improve the dietary health of infants and toddlers while ensuring their proper care.

[0069] In one embodiment, please refer to Figure 4 S3 includes:

[0070] S31: When recognizing infants and young children eating, extract the target images corresponding to each frame of the infants and young children eating;

[0071] Specifically, since the input is a multi-frame image decomposed from a real-time video stream, some images show infants eating, while others do not. The target images of the frames where infants are identified as eating are extracted as the target images.

[0072] S32: Based on the hand position information of the person and the mouth position information of the infant, the number of times the infant eats is calculated;

[0073] Specifically, the number of times the infants eat is counted for each frame of the target image: a number of times the infants eat is pre-set as N1, with an initial value of 0. If the absolute value of the difference between x2 and x4 is less than 5 and the absolute value of the difference between y2 and y4 is less than 4, then N1 is incremented by 1. The final N1 is then counted as the number of times the infants eat.

[0074] S33: Based on the preset meal quality assessment rules and combined with the number of times the infants and young children eat, a comprehensive analysis is performed on each of the target images to assess the meal quality of the infants and young children.

[0075] Specifically, extracting target images showing infants' eating behaviors for behavioral analysis, without processing the remaining images where no infant eating behaviors were identified, can effectively improve the efficiency of eating behavior quality assessment, reduce workflow, and save time.

[0076] In one embodiment, please refer to Figure 5 S33 includes:

[0077] S331: Perform an autonomy analysis on the infant's eating behavior in the target image, and obtain an autonomy score by combining the autonomy analysis results with the number of times the infant eats;

[0078] In one embodiment, please refer to Figure 6 S331 includes:

[0079] S3311: A first classification model is pre-set, and the human hand is classified into infant hand and non-infant hand using the first classification model;

[0080] Specifically, in the large number of images collected in the infant care scenario, the infant's hand and the non-infant's hand in these images are pre-labeled. The images are then trained using a deep learning algorithm to output a first classification model based on ResNet that can classify the hands in the images into infant's hands and non-infant's hands. Using the first classification model, the person's hand is classified into infant's hands and non-infant's hands.

[0081] S3312: Obtain the number of times the infant's hands appear, and count the number of times the infant eats independently based on the number of times the infant's hands appear;

[0082] Specifically, please see Figure 7 If an infant's hand appears as labeled 5, it is considered that the infant puts the food into their mouth by themselves, rather than being fed by a guardian or other adult. In other words, the infant is feeding themselves. A number of times the infant feeds themselves is preset, N2. The initial value of N2 is 0. If the infant's hand appears once during the feeding process, N2 is incremented by 1. The final N2 is counted as the number of times the infant feeds themselves.

[0083] S3313: The autonomy score is calculated by combining the number of times the infant eats and the number of times the infant eats independently.

[0084] Specifically, the autonomy score P1 is calculated using the number of times the infant eats (N1) and the number of times the infant eats independently (N2), where 0 < P1 < 10. The formula is: P1 = N2 / N1 * 10. A higher P1 value indicates higher autonomy in the infant's eating, while a lower P1 value indicates lower autonomy, suggesting the infant may be mostly fed by others. A higher P1 value indicates higher autonomy, suggesting the infant may mostly eat independently, indicating a higher appetite. Effective evaluation of the infant's eating autonomy can remind parents to reduce feeding when autonomy is low and encourage the infant to develop good independent eating habits, which is beneficial to the infant's physical and mental health development.

[0085] S332: Perform a pleasure analysis on the infants' eating behavior in the target image, and obtain a pleasure score by combining the pleasure analysis results with the number of times the infants eat;

[0086] In one embodiment, please refer to Figure 8 S332 includes:

[0087] S3321: Perform face detection on the target image to identify the facial information of the infant in the target image;

[0088] Specifically, in a large number of images collected from infant care scenarios, images of infants' faces are pre-labeled, and deep learning algorithms are used to train the model to output a third detection model that can detect infants' faces. The target image is then input into the third detection model to identify the infant's facial information in the target image. The infant's facial information includes at least one of the following: key point information such as the left eye, right eye, nose, and mouth.

[0089] S3322: A second classification model is pre-set, and the infant's facial information is input into the second classification model to identify the infant's facial expression information;

[0090] Specifically, in a large number of images collected from infant care scenarios, key points of the infants are pre-labeled when they are crying, smiling, and in other states. The data is then trained using a deep learning algorithm to output a second classification model that can classify the infants' facial information into smiling, crying, or other states. Using the second classification model, the infants' facial information is classified into smiling, crying, and other states, and the information of the infants in smiling and crying states is taken as the infants' facial expression information.

[0091] S3323: Based on the aforementioned facial expression information, count the number of times the infant smiled and cried;

[0092] Specifically, a number of infant smiles M1 is preset, with an initial value of 0. If an expression indicating an infant is smiling appears, M1 is incremented by 1, and the final M1 is counted as the number of infant smiles. Similarly, a number of infant cries M2 is preset, with an initial value of 0. If an expression indicating an infant is crying appears, M2 is incremented by 1, and the final M2 is counted as the number of infant cries.

[0093] S3324: Calculate the pleasure score by combining the number of times the infant smiled, cried, and ate.

[0094] Specifically, the pleasure score P2 is calculated using the number of times the infant smiles (M1), cries (M2), and eats (N1) of the infant. The formula is: P2 = (M1 - M2) / N1 * 10. If the pleasure score -10 = < P2 < 0, it indicates that the infant is crying during mealtime and has a low level of pleasure. If the pleasure score P2 >= 0, and the larger the value of P2, the more likely the infant is smiling during mealtime and has a high level of pleasure. By detecting changes in the infant's facial expressions during mealtime, the pleasure level of the infant can be effectively monitored. When the pleasure level is low, it may be due to picky eating or food that is too hot, requiring parents to appropriately adjust the type and temperature of the food.

[0095] S333: Perform compactness analysis on the infant and toddler eating behavior in the target image, and obtain a compactness score by combining the compactness analysis results with the number of times the infant and toddler eats.

[0096] In one embodiment, please refer to Figure 9 S333 includes:

[0097] S3331: Obtain the timing of infant feedings;

[0098] Specifically, when an infant's feeding is detected, the meal quality assessment device will automatically record the time points as T1, T2, ... T(N1), where T1, T2, ... T(N1) represent the time points from the infant's first feeding to the last N1th feeding.

[0099] S3332: Based on the stated time points, calculate the average interval time and time series;

[0100] Specifically, based on the aforementioned time points, the average interval time Tavg between infant feedings is calculated using the formula: Tavg = (T(N1) - T1) / (N1 - 1). Then, the time sequence T(i,j) between two adjacent infant feedings is calculated, where 1 < i, j < N1, and i is an integer. The formula is as follows: T(1,2) = T2 - T1, T(2,3) = T3 - T2, ..., T(N1 - 1, N1) = T(N1) - T(N1 - 1).

[0101] S3333: Based on the average interval time, time series and the number of times the infants and young children eat, calculate the compactness score.

[0102] Specifically, a compactness count num is pre-defined, with an initial value of 0. If T(i,j) is less than or equal to Tavg, it indicates that the time interval between two consecutive feedings by the infant is less than the average time interval. This suggests that the infant's mealtime is relatively compact and they are quite active in eating. Therefore, num is incremented by 1 to obtain the final compactness count num. Based on the compactness count num and the number of feedings N1, the compactness score P3 is calculated using the following formula: P3 = num / (N1-1)*10, where 0 < P3 < 10. The larger the compactness score P3, the more compact the infant's mealtime and the higher their enthusiasm for eating; conversely, the smaller the P3 value, the less compact the infant's mealtime and the lower their enthusiasm for eating, requiring parents to encourage the infant to eat more actively.

[0103] S334: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score. Based on the comprehensive score, the evaluation result is given.

[0104] In one embodiment, please refer to Figure 10 S334 includes:

[0105] S3341: Obtain the age of the infant or toddler and pre-set the score threshold and age threshold;

[0106] Specifically, the device obtains the infant's age Age input by the user through a mobile terminal, and pre-sets a score threshold P threshold, taking P threshold = 6 as an example, and pre-sets an age threshold Age threshold for the infant, taking age threshold Age threshold = 3 as an example.

[0107] S3342: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score;

[0108] Specifically, the autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score. The calculation formula is: P = a × P1 + b × P2 + c × P3, where a, b, and c are the weighting coefficients of the three scores, 0 < a, b, and c < 1, and a + b + c = 1. Considering that different parents have different requirements for assessing the quality of infants' and toddlers' meals, the three weighting coefficients a, b, and c can be set by the user to meet different user needs. For example, if the user wants to prefer assessing the autonomy of infants and toddlers' meals, the value of the weighting coefficient a is set to 0.6. Similarly, if the user wants to prefer assessing pleasure and compactness, the weighting coefficients b and c are increased respectively.

[0109] S3343: Compare the overall score with the score threshold, and compare the infant's age with the age threshold to obtain the assessment result of the infant's meal quality.

[0110] Specifically, taking a weighting coefficient of a = 0.6, b = 0.3, and c = 0.1 set by the user as an example, in this case, the user prefers to assess the infant's autonomy in eating, and the calculated comprehensive score P is 5, which is less than the preset score threshold P. The infant's age is 4, which is greater than the preset age threshold, indicating that the infant is older but has poor self-feeding ability. Parents need to reduce the frequency of feeding the infant and encourage them to eat independently more often. If the user sets a weighting coefficient of a = 0.1, b = 0.6, and c = 0.3, in this case, the user prefers to assess the infant's enjoyment during meals, and the calculated comprehensive score P is 4, which is less than the preset score threshold P. The P-threshold indicates poor enjoyment during mealtimes for infants and toddlers. Parents need to pay attention to the reasons for this low enjoyment, such as picky eating or unsuitable food temperature, and adjust the type and temperature of the food accordingly. If the user sets the weighting coefficients a=0.1, b=0.3, and c=0.6, the user prefers to assess the intensity of the infant's mealtime. The calculated comprehensive score P is 5, which is less than the preset score threshold P. In this case, the infant's age Age is 5, which is greater than the set age threshold Age, indicating that the infant is older but has low mealtime intensity. Parents need to improve the infant's attention during mealtimes and maintain consistent mealtimes. The comprehensive score presents the user with specific evaluation results. The entire process is intelligent and digital, which assists in guiding infants' mealtime behavior and helps improve the standardization of the infant's mealtime process.

[0111] Example 2

[0112] Please see Figure 11 Embodiment 2 of the present invention also provides an intelligent assessment device for the quality of infant and toddler meals, the device comprising:

[0113] The image acquisition module is used to acquire real-time video streams in infant and toddler care scenarios and decompose the video streams into multiple frames of images.

[0114] The behavior determination module is used to pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hands and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hands and the position information of the infant's mouth.

[0115] The behavior assessment module is used to assess the eating behavior of infants and toddlers based on preset behavior analysis rules when they are identified eating, and output the assessment results.

[0116] Specifically, the intelligent assessment device for infant and toddler meal quality according to Embodiment 2 of this invention includes an image acquisition module for acquiring real-time video streams in infant and toddler care scenarios and decomposing the video streams into multiple frames; a behavior determination module for pre-setting a target detection model, inputting the multiple frames into the target detection model, identifying the position information of a person's hands and the position information of the infant's mouth in the multiple frames, and identifying whether the infant is eating based on the position information of the person's hands and the position information of the infant's mouth; and a behavior assessment module for assessing the infant's eating behavior according to preset behavior analysis rules when the infant is identified as eating, and outputting the assessment results. When infant and toddler eating behavior occurs, the preset behavior analysis rules provide a more comprehensive analysis and assessment of the infant's eating behavior, resulting in detailed assessment results. This helps users guide infants and toddlers in their meals, improves the standardization of infants' and toddlers' meals, and also enhances the care experience for parents.

[0117] Example 3

[0118] In addition, combined Figure 1 The intelligent assessment method for infant and toddler meal quality described in Embodiment 1 of the present invention can be implemented by an electronic device. Figure 12 A schematic diagram of the hardware structure of the electronic device provided in Embodiment 3 of the present invention is shown.

[0119] Electronic devices may include processors and memory storing computer program instructions.

[0120] Specifically, the processor may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement embodiments of the present invention.

[0121] The memory may include a large-capacity storage device for data or instructions. For example, and not limitingly, the memory may include a hard disk drive (HDD), a floppy disk drive, flash memory, an optical disk drive, a magneto-optical disk drive, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, the memory may include removable or non-removable (or fixed) media. Where appropriate, the memory may be internal or external to a data processing device. In a particular embodiment, the memory is a non-volatile solid-state memory. In a particular embodiment, the memory includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically rewritable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0122] The processor reads and executes computer program instructions stored in the memory to implement any of the intelligent assessment methods for infant and toddler meal quality in the above embodiments.

[0123] In one example, the electronic device may also include a communication interface and a bus. For example, Figure 12 As shown, the processor, memory, and communication interface are connected via a bus and communicate with each other.

[0124] The communication interface is mainly used to enable communication between various modules, devices, units and / or equipment in the embodiments of the present invention.

[0125] A bus, including hardware, software, or both, couples components of the device together. For example, and not limitingly, a bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, a bus may include one or more buses. While specific buses are described and illustrated in embodiments of the invention, the invention contemplates any suitable bus or interconnect.

[0126] Example 4

[0127] In addition, in conjunction with the intelligent assessment method for infant and toddler meal quality in Embodiment 1 above, Embodiment 4 of the present invention can also provide a computer-readable storage medium for implementation. This computer-readable storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any one of the intelligent assessment methods for infant and toddler meal quality in the above embodiments.

[0128] In summary, the embodiments of the present invention provide an intelligent assessment method, device, equipment, and storage medium for infants' and toddlers' meal quality.

[0129] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0130] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the required tasks. The programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0131] It should also be noted that the exemplary embodiments mentioned in this invention describe methods or systems based on a series of steps or apparatus. However, this invention is not limited to the order of the steps described above; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0132] The above description is merely a specific embodiment of the present invention. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the protection scope of the present invention.

Claims

1. A method for intelligently assessing the quality of infants' and toddlers' meals, characterized in that, The method includes: S1: Acquire a real-time video stream in an infant care scenario and decompose the video stream into multiple frames of images; S2: Pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hand and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hand and the position information of the infant's mouth. S3: When an infant or toddler is identified eating, the eating behavior of the infant or toddler is analyzed according to the preset behavior analysis rules to evaluate the quality of the infant or toddler's meal. S3 includes: S31: When recognizing infants and young children eating, extract the target images corresponding to each frame of the infants and young children eating; S32: Based on the hand position information of the person and the mouth position information of the infant, the number of times the infant eats is calculated; S33: Based on the preset meal quality assessment rules and combined with the number of times the infants and young children eat, a comprehensive analysis is performed on each of the target images to assess the meal quality of the infants and young children. S33 includes: S331: Perform an autonomy analysis on the infant's eating behavior in the target image, and obtain an autonomy score by combining the autonomy analysis results and the number of times the infant eats. The autonomy score is calculated by the ratio of the number of times the infant eats to the number of times the infant eats voluntarily. S332: Perform a pleasure analysis on the infant's eating behavior in the target image, and obtain a pleasure score by combining the pleasure analysis results and the number of times the infant eats. The pleasure score is calculated by combining the number of times the infant smiles, cries, and eats. S333: Perform compactness analysis on the infant eating behavior in the target image, and obtain a compactness score by combining the compactness analysis results and the number of times the infant eats. The compactness score is calculated based on the average interval between infants' eating, the time sequence between two adjacent infants' eating, and the number of times the infant eats. S334: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score. Based on the comprehensive score, the evaluation result is given.

2. The intelligent assessment method for infant and toddler meal quality according to claim 1, characterized in that, S331 includes: S3311: A first classification model is pre-set, and the human hand is classified into infant hand and non-infant hand using the first classification model; S3312: Obtain the number of times the infant's hands appear, and count the number of times the infant eats independently based on the number of times the infant's hands appear; S3313: The autonomy score is calculated by combining the number of times the infant eats and the number of times the infant eats independently.

3. The intelligent assessment method for infant and toddler meal quality according to claim 1, characterized in that, S332 includes: S3321: Perform face detection on the target image to identify the facial information of the infant in the target image; S3322: A second classification model is pre-set, and the infant's facial information is input into the second classification model to identify the infant's facial expression information; S3323: Based on the aforementioned facial expression information, count the number of times the infant smiled and cried; S3324: Calculate the pleasure score by combining the number of times the infant smiled, cried, and ate.

4. The intelligent assessment method for infant and toddler meal quality according to claim 1, characterized in that, S333 includes: S3331: Obtain the timing of infant feedings; S3332: Based on the stated time points, calculate the average interval time and time series; S3333: Based on the average interval time, time series and the number of times the infants and young children eat, calculate the compactness score.

5. The intelligent assessment method for infant and toddler meal quality according to claim 1, characterized in that, S334 includes: S3341: Obtain the age of the infant or toddler and pre-set the score threshold and age threshold; S3342: The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score; S3343: Compare the overall score with the score threshold, and compare the infant's age with the age threshold to obtain the assessment result of the infant's meal quality.

6. An intelligent assessment device for the quality of infant and toddler meals, characterized in that, The device includes: The image acquisition module is used to acquire real-time video streams in infant and toddler care scenarios and decompose the video streams into multiple frames of images. The behavior determination module is used to pre-set a target detection model, input the multi-frame images into the target detection model, identify the position information of the human's hands and the position information of the infant's mouth in the multi-frame images, and identify whether the infant is eating based on the position information of the human's hands and the position information of the infant's mouth. The behavior assessment module is used to analyze the eating behavior of infants and toddlers based on preset behavior analysis rules when they are identified eating, and to assess the quality of their meals. When an infant or toddler is identified eating, the process of analyzing their eating behavior according to preset behavioral analysis rules and evaluating the quality of their meal includes: When identifying infants and toddlers eating, extract the target images corresponding to each frame of the infant or toddler eating; Based on the hand position information of the person and the mouth position information of the infant, the number of times the infant ate was calculated. Based on the preset meal quality assessment rules and combined with the number of times the infants and young children eat, the target images are comprehensively analyzed to assess the meal quality of the infants and young children. The assessment of infants' meal quality, based on preset meal quality evaluation rules and the frequency of infant feedings, involves a comprehensive analysis of the target images to evaluate the infants' meal quality. An autonomy analysis is performed on the infants' eating behavior in the target image. An autonomy score is obtained by combining the autonomy analysis results with the number of times the infants eat. The autonomy score is calculated by the ratio of the number of times the infants eat to the number of times the infants eat autonomously. A pleasure analysis is performed on the infants' eating behavior in the target image. A pleasure score is obtained by combining the pleasure analysis results with the number of times the infants eat. The pleasure score is calculated by combining the number of times the infants smile, cry, and eat. A compactness analysis is performed on the infants' eating behavior in the target image. The compactness analysis results and the number of times the infants eat are combined to obtain a compactness score. The compactness score is calculated based on the average interval between infants' eating, the time sequence between two adjacent infants' eating, and the number of times the infants eat. The autonomy score, pleasure score, and compactness score are weighted and calculated to obtain a comprehensive score. Based on the comprehensive score, the evaluation result is given.

7. An electronic device, characterized in that, include: At least one processor, at least one memory, and computer program instructions stored in the memory, which, when executed by the processor, implement the method as described in any one of claims 1-5.

8. A storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by a processor, the method as described in any one of claims 1-5 is implemented.

Images