Imprint collector and visualization system thereof

Abstract

The present invention discloses an impression collector and a visualization system thereof, and relates to the technical field of impression collection. The invention collects initial impression images, performs impression analysis and processing through an image processing unit, evaluates the impression collection accuracy and analyzes the health status of newborns' vital signs, and then performs early warning prompts and visualization presentation through a data display unit, thereby continuously improving the impression collection accuracy and ensuring the stability of the impression collector. The invention is suitable for archiving and registering the footprints of newborns after pregnancy, and collecting and analyzing footprints through regular physical examinations of newborns, which can assist in medical diagnosis and health assessment of infants and young children, and understand their growth and development and possible abnormalities.

Description

Technical Field

[0001] The present invention relates to the technical field of impression collection, and in particular to an impression collector and a visualization system thereof. Background Art

[0002] In the field of medical health, footprints can reflect the biological characteristics of newborns, such as the skin, arch and toe bones of the newborn. Regular collection, measurement and comparison of newborn footprints can help us understand the growth and development of newborns, as well as their recent movement behavior, standing posture and other information. Therefore, footprints can be used as a health reference indicator to monitor the growth and development of newborns.

[0003] After a newborn is born, it is necessary to collect footprints and archive them in the maternal medical history. However, the existing method of collecting footprints is generally in the form of manual stamping by nurses. Due to the universal development of electronic medical record systems, the manual stamping method is difficult to be used for information-based archiving, and there may be problems such as blurred footprints and poor accuracy of footprint collection. In addition, the manual archiving mode is difficult to perform automated health analysis of newborns through footprints.

[0004] In view of the above technical defects, a solution is now proposed. Summary of the invention

[0005] The purpose of the present invention is to solve the problems of low precision, blurred traces, and difficulty in information archiving in the existing manual footprint collection mode, as well as the defect that it is difficult to perform automated health analysis of newborns through footprint traces. The footprint image of the newborn is collected by the footprint collection unit based on image scanning technology, and then the footprint analysis is processed by the image processing unit to evaluate the footprint collection accuracy and analyze the physical health status of the newborn. Then, the data display unit is used to provide early warning prompts and visual presentation, thereby continuously improving the footprint collection accuracy and ensuring the stability of the footprint collector. The invention is suitable for archiving and registering the footprints of newborns after pregnancy, and collecting and analyzing footprints through regular physical examinations of newborns. It can assist in medical diagnosis and health assessment of infants and young children, and understand growth and development conditions and possible abnormalities.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] A print collection visualization system includes a print collection unit, an image processing unit and a data display unit, wherein the print collection unit, the image processing unit and the data display unit are connected by signals, wherein the image processing unit includes a preprocessing module and a depth analysis module, and the preprocessing module and the depth analysis module are connected by signals;

[0008] The impression collection unit is used to collect at least one initial impression image and store it in the maternal electronic medical record, and then upload the initial impression image to the image processing unit;

[0009] The image processing unit is used to analyze and process the initial printed image:

[0010] The grayscale enhancement processing of the initial print image is performed through the preprocessing module to generate an imprint image, and the initial print image and the imprint image are compared to analyze and evaluate the imprint acquisition accuracy and generate an early warning signal;

[0011] The deep analysis module then analyzes the texture features of the footprint image to assess the abnormalities of the skin surface and bone development of the footprint, and then comprehensively analyzes and assesses the physical health of the newborn, and generates prompt signals and footprint health reports;

[0012] The data display unit is used to receive signals and footprint health reports, and make corresponding visual presentations for reference by professional medical staff.

[0013] Furthermore, the specific process of the impression analysis is as follows:

[0014] The grayscale enhancement process is performed on the initial printing image through the pre-processing module, so as to obtain the print image;

[0015] By comparing the initial print image with the print image, the print acquisition accuracy coefficient is obtained, so as to analyze and evaluate the print acquisition accuracy and generate corresponding warning signals;

[0016] Then, the texture feature analysis of the print image is performed through the depth analysis module to obtain the skin abnormality coefficient and the developmental abnormality coefficient, and to evaluate the skin surface abnormality and bone developmental abnormality of the print;

[0017] By combining the skin abnormality coefficient and the developmental abnormality coefficient, a comprehensive health assessment index is obtained to evaluate the physical health of the newborn, and prompt signals and footprint health reports are generated for visual presentation.

[0018] Furthermore, the specific process of grayscale enhancement processing is as follows:

[0019] The first printed image is marked as P0, and the image P0 is preset to have n0 feature points. Any feature point is marked as a. A regular quadrilateral grid area is delineated with the feature point a as the center, and the side length of the regular quadrilateral grid is preset to be d, so as to obtain and mark it as a grid Wa;

[0020] There are n1 points in the preset grid Wa. The grayscale values ​​of all points in the grid Wa are acquired through image processing software, and any point in the grid Wa is marked as b, and the grayscale value of point b is marked as Hb. A discrete distribution graph is established for data display.

[0021] First, obtain the median ZWa of the grayscale values ​​of n1 points in the grid Wa, and obtain the average PJa of the grayscale values ​​of n1 points;

[0022] Through the median ZWa and average PJa of the grayscale value, a formula is established to obtain the grayscale conversion standard value Ha;

[0023] Then, the standard deviation σb of the gray value is obtained, and the kurtosis coefficient FDb of the gray value and the skewness coefficient PDb of the gray value are calculated;

[0024] The grayscale difference contrast value Zcy is obtained by combining the standard deviation σb, the kurtosis coefficient FDb and the skewness coefficient PDb of the grayscale value;

[0025] The threshold F0 of the grayscale difference contrast value Zcy is set, and then the threshold comparison is performed to transform and replace the grayscale values ​​of all points of the grid Wa. After the grayscale data of the grid area where the n0 feature points of the image P0 are located is transformed and adjusted in turn, an imprint image is generated and marked as P1.

[0026] Furthermore, the specific process of transformation and replacement is:

[0027] If the grayscale difference contrast value Zcy is less than the threshold F0, the grid Wa is transformed as a whole, and the grayscale values ​​of all points of the grid Wa are adjusted to the grayscale transformation standard value Ha;

[0028] If the grayscale difference contrast value Zcy ≥ threshold F0, the grayscale values ​​of all points of the grid Wa are compared with the grayscale conversion standard value Ha in turn: if the grayscale value of point b is lower than or equal to the grayscale conversion standard value Ha, no processing is performed; if the grayscale value of point b is higher than the grayscale conversion standard value Ha, the grayscale value of point b is enhanced, specifically: set the grayscale enhancement range φ%, obtain the grayscale enhancement value Hd through the grayscale enhancement range φ%, and adjust the grayscale value of point b to Hd.

[0029] Furthermore, the specific process of analyzing and evaluating the accuracy of imprint collection is as follows:

[0030] Place the initial print image P0 and the imprint image P1 in the same coordinate system, extract m1 points of the initial print image P0, mark any point as e, and mark the point of the imprint image P1 with the same coordinate as the point as f;

[0031] The gray value of point e is marked as He, and the gray value of point f is marked as Hf, and then the gray difference coefficient Xh is obtained;

[0032] Then, the grayscale difference coefficient of point m1 is used to comprehensively obtain the print collection accuracy coefficient Xyh;

[0033] Set the evaluation interval of the impression collection accuracy coefficient Xyh, evaluate the impression collection accuracy through interval comparison and generate corresponding warning signals:

[0034] There are R evaluation intervals preset, and any evaluation interval is marked as Qr, where r is the serial number of the evaluation interval, 0<r≤R. When the print collection accuracy coefficient Xyh is in the evaluation interval Qr, an r-level warning signal is generated, the print collection accuracy is evaluated to be r-level, and it is output to the data display unit for visual presentation.

[0035] Furthermore, the specific process of texture feature analysis is as follows:

[0036] First, a gray-level co-occurrence matrix is ​​constructed, and N0 groups of neighboring pixel pairs are extracted. Any group of neighboring pixel pairs is marked as G. The horizontal direction is used as the reference direction. The neighboring pixel pair G includes pixel g1 and pixel g2. The gray value of any pixel g1 and its adjacent pixel g2 are compared to determine the gray difference Δg of the neighboring pixel pair G.

[0037] Classify the N0 groups of neighboring pixel pairs according to the grayscale difference Δg, accumulate the frequency Jg of the neighboring pixel pairs with the grayscale difference Δg in the N0 groups of neighboring pixel pairs, and then obtain the grayscale difference Δgm corresponding to the highest frequency Jg by sorting in descending order;

[0038] A grayscale difference threshold φg is preset. When the grayscale difference Δg is higher than the grayscale difference threshold φg, the group of neighboring pixel pairs is marked as YCg, and the pixels with high grayscale values ​​in the neighboring pixel pairs YCg are extracted and marked with characteristic colors. Then, through comparative analysis of N0 groups of neighboring pixel pairs, all pixel points Co marked with characteristic colors are obtained, and the closed area formed by connecting the pixel points Co is marked as the characteristic texture area.

[0039] Perform grayscale co-occurrence matrix analysis on the characteristic texture area to obtain the grayscale difference Δgz corresponding to the characteristic texture area, set the floating interval Qd[Δgm-z, Δgm+z] of the grayscale difference Δgz, and when the grayscale difference Δgz exceeds the floating interval Qd, the characteristic texture area is determined to be a skin abnormality area;

[0040] Then, the number and area of ​​abnormal skin regions are measured and marked as Np and Sp respectively, and then the skin abnormality coefficient Xpf is obtained by combining the number Np and area Sp of abnormal skin regions;

[0041] According to professional medical knowledge, the print image P1 is partitioned into foot structures to obtain the arch area and the toe area: if the characteristic texture area is located in the arch area, it is marked as the arch area Z0 and an arch abnormality analysis is performed; if the characteristic texture area is located in the toe area, it is marked as the toe area Z1 and an toe abnormality analysis is performed;

[0042] The arch area deviation Pg is obtained through arch abnormality analysis, the toe area deviation Pz is obtained through toe abnormality analysis, and then the developmental abnormality coefficient Xfy is obtained by combining the arch area deviation Pg and the toe area deviation Pz.

[0043] Furthermore, the specific process of arch abnormality analysis and toe abnormality analysis is as follows:

[0044] Arch abnormality analysis: Set the arch area standard proportion ZG, calculate the arch standard area Sa0 of the imprint image P1 according to the footprint area of ​​the imprint image P1, and measure the actual area Sa1 of the arch area Z0 to obtain the arch area deviation Pg;

[0045] Toe abnormality analysis: Set the toe area standard proportion ZB, calculate the toe standard area Sb0 of the print image P1 based on the footprint area of ​​the print image P1, and measure the actual area Sb1 of the toe area Z1 to obtain the toe area deviation Pz.

[0046] Furthermore, the specific process of assessing the physical health status of the newborn is as follows:

[0047] By combining the skin abnormality coefficient Xpf and the developmental abnormality coefficient Xfy, the health assessment index JK is comprehensively obtained;

[0048] Set the standard interval of the health assessment index JK, evaluate the health of the newborn's physical signs through interval comparison, and generate prompt signals and footprint health reports for visual presentation:

[0049] There are U standard intervals preset, and any standard interval is marked as Qu, where u is the serial number of the standard interval, 0<u≤U. When the health assessment index JK is in the standard interval Qu, a u-level prompt signal is generated, and the physical health level of the newborn is assessed as u-level, and it is output to the data display unit for visual presentation;

[0050] The skin abnormality coefficient Xpf, developmental abnormality coefficient Xfy and health assessment index JK are then integrated to generate a footprint health report, which is then output to a data display unit for visual presentation for reference by professional medical staff.

[0051] Furthermore, the initial printed image is a footprint image or a handprint image.

[0052] An impression collector comprises the impression collection visualization system as described in any one of claims 1 to 9.

[0053] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0054] The present invention collects the initial print image through the print collection unit, and then performs print analysis and processing through the image processing unit, evaluates the print collection accuracy and analyzes the health status of the newborn's physical signs, and then performs early warning prompts and visual presentation through the data display unit, thereby continuously improving the print collection accuracy and ensuring the stability of the print collector. It is suitable for archiving and registering the footprints of newborns after pregnancy, and collecting and analyzing the footprints through regular physical examinations of newborns, which can assist in medical diagnosis and health assessment of infants and young children, and understand the growth and development status and possible abnormalities.

[0055] The present invention enhances the contrast of the initial print image through grayscale enhancement processing, and obtains a refined impression image for subsequent texture feature analysis. The skin abnormalities on the footprint surface and the bone abnormalities in the toes and arches of the footprint are analyzed through the impression texture features, so as to comprehensively evaluate the health status of the newborn represented by the footprint trace, so as to provide early warning of potential growth and development health problems of the newborn. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Shows a schematic diagram of the modules of the present invention;

[0057] Figure 2 A schematic diagram of the process of the present invention is shown. DETAILED DESCRIPTION

[0058] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0059] Embodiment 1:

[0060] like Figure 1-2 As shown, an impression collector, using the impression collection visualization system described below, includes an impression collection unit, an image processing unit and a data display unit, wherein the image processing unit includes a preprocessing module and a depth analysis module; the impression collection unit, the image processing unit and the data display unit are signal connected, and the preprocessing module and the depth analysis module are signal connected;

[0061] The working steps are as follows:

[0062] Sa, the footprint collection unit collects the initial footprint image and stores it in the maternal electronic medical record, and then uploads the newborn's footprint image to the image processing unit;

[0063] The initial print image is a newborn's footprint image, handprint image, or a combination of the parents' handprint image and footprint image. Usually, the newborn's footprint image and the parents' handprint image form a label image, which can be subsequently printed and stored as a souvenir.

[0064] Sb, the image processing unit performs an impression analysis on the initial print image: the grayscale enhancement processing is performed on the initial print image through the preprocessing module to obtain the impression image, and the initial print image and the impression image are compared to analyze and evaluate the impression acquisition accuracy and generate an early warning signal; the texture feature analysis of the initial print image is performed through the depth analysis module to evaluate the skin surface abnormality and bone development abnormality of the impression, and then comprehensively analyze and evaluate the physical health status of the newborn, and generate a prompt signal and a footprint health report. The specific process of the impression analysis processing is as follows:

[0065] Sb-1, grayscale enhancement processing is performed on the initial printed image through the pre-processing module to obtain the imprint image. The specific process of grayscale enhancement processing is as follows:

[0066] Sb-101, mark the first printed image as P0, presume that the image P0 has n0 feature points, mark any feature point as a, delineate a regular quadrilateral mesh area with the feature point a as the center, and presume that the side length of the regular quadrilateral mesh is d, thereby obtaining and marking it as mesh Wa;

[0067] Sb-102, there are n1 points in the preset grid Wa, the grayscale values ​​of all points in the grid Wa are acquired through image processing software, and any point in the grid Wa is marked as b, and the grayscale value of point b is marked as Hb, and a discrete distribution graph is established for data display;

[0068] Sb-103, first obtain the median ZWa of the grayscale values ​​of n1 points in the grid Wa, and obtain the average PJa of the grayscale values ​​of n1 points:

[0069] Sb-104, through the median ZWa and average PJa of the grayscale value, establish a formula to obtain the grayscale conversion standard value Ha: Ha = ZWa α1 +PJa α2 ;

[0070] Among them, α1 and α2 are weight factor coefficients of the median ZWa and average PJa of the grayscale value, respectively, and both α1 and α2 are greater than 0; the weight factor coefficients are preset through a large amount of data experimental measurement, and when the median ZWa and average PJa of the grayscale value are higher, the grayscale conversion standard value Ha is higher;

[0071] Sb-105, and then find the standard deviation of the gray value σb: And calculate the kurtosis coefficient FDb of the gray value and the skewness coefficient PDb of the gray value:

[0072] Sb-105-1, kurtosis coefficient FDb of gray value:

[0073]

[0074] When the kurtosis coefficient FDb of the gray value is higher, it means that the distribution of the gray values ​​of n1 points in the grid Wa is steeper, indicating that the overall gray value difference in the grid area is greater; when the kurtosis coefficient FDb of the gray value is lower, it means that the distribution of the gray values ​​of n1 points in the grid Wa is flatter, indicating that the overall gray values ​​in the grid area tend to be consistent, where steepness and flatness are relative to the normal distribution. When the kurtosis coefficient FDb of the gray value is 0, it means that the distribution of the gray values ​​of n1 points in the grid Wa shows a normal distribution trend;

[0075] Sb-105-2, skewness coefficient PDb of gray value:

[0076]

[0077] When the skewness coefficient PDb of the gray value is greater than 0, it means that the average gray value PJa of n1 points in the grid Wa is higher than the mode of the gray value, indicating that the gray value distribution of n1 points is right-skewed, and the higher the value of the skewness coefficient PDb, the higher the degree of the right-skewed trend; when the skewness coefficient PDb of the gray value is less than 0, it means that the average gray value PJa of n1 points in the grid Wa is lower than the mode of the gray value, indicating that the gray value distribution of n1 points is left-skewed, and the lower the value of the skewness coefficient PDb, the higher the degree of the left-skewed trend; when the kurtosis coefficient FDb of the gray value is closer to 0, it means that the average gray value PJa of n1 points in the grid Wa is closer to the mode, which means that the gray value distribution of n1 points is more uniform;

[0078] Sb-106, by combining the standard deviation σb, kurtosis coefficient FDb and skewness coefficient PDb of the gray value, obtain the gray difference contrast value Zcy: Zcy=ω1*σb+ω2*FDb+ω3*PDb 2 ;

[0079] Among them, ω1, ω2 and ω3 are weight factor coefficients of the standard deviation σb, kurtosis coefficient FDb and skewness coefficient PDb of the gray value, respectively, and ω1, ω2 and ω3 are all greater than 0; when the standard deviation σb and kurtosis coefficient FDb of the gray value are higher, and the square of the skewness coefficient PDb is larger, the gray difference contrast value Zcy is higher, indicating that the gray value distribution difference of n1 points in the grid Wa area is greater;

[0080] Sb-107, set the threshold F0 of the grayscale difference contrast value Zcy, and then perform threshold comparison to transform and replace the grayscale values ​​of all points of the grid Wa:

[0081] If the grayscale difference contrast value Zcy is less than the threshold F0, the grid Wa is transformed as a whole, and the grayscale values ​​of all points of the grid Wa are adjusted to the grayscale transformation standard value Ha;

[0082] If the grayscale difference comparison value Zcy ≥ threshold F0, the grayscale values ​​of all points of the grid Wa are compared with the grayscale conversion standard value Ha in turn: if the grayscale value of point b is lower than or equal to the grayscale conversion standard value Ha, no processing is performed; if the grayscale value of point b is higher than the grayscale conversion standard value Ha, the grayscale value of point b is enhanced, specifically: set the grayscale enhancement range φ%, obtain the grayscale enhancement value Hd through the grayscale enhancement range φ%, and adjust the grayscale value of point b to Hd;

[0083] Sb-108, by converting and adjusting the grayscale data of the grid area where the n0 feature points of the image P0 are located, an imprint image is generated and marked as P1;

[0084] Sb-2, and obtain the print acquisition accuracy coefficient by comparing the initial print image with the print image, so as to analyze and evaluate the print acquisition accuracy and generate corresponding warning signals. The specific process of analyzing and evaluating the print acquisition accuracy is as follows:

[0085] Sb-201, place the initial print image P0 and the impression image P1 in the same coordinate system, extract m1 points of the initial print image P0, mark any point as e, and mark the point of the impression image P1 with the same coordinate as the point as f;

[0086] Sb-202, mark the gray value of point e as He, and mark the gray value of point f as Hf, and then obtain the gray difference coefficient Xh:

[0087] Sb-203, when the difference between the grayscale value He of point e and the grayscale value Hf of point f is higher, the grayscale difference coefficient Xh is higher, and then the grayscale difference coefficient of point m1 is used to comprehensively obtain the print collection accuracy coefficient Xyh:

[0088] Wherein, μ is the conversion coefficient and μ is greater than 0; the conversion coefficient μ is a constant value of the print acquisition accuracy coefficient converted by combining the grayscale difference coefficients of m1 points, and the conversion coefficient μ is obtained by preset; when the grayscale difference coefficient Xh is higher, the print acquisition accuracy coefficient Xyh is lower, indicating that the accuracy of print acquisition is worse;

[0089] Sb-204, set the evaluation interval of the print collection accuracy coefficient Xyh, evaluate the print collection accuracy through interval comparison and generate corresponding warning signals:

[0090] There are R evaluation intervals preset, and any evaluation interval is marked as Qr, where r is the serial number of the evaluation interval, 0<r≤R, and when the print collection accuracy coefficient Xyh is in the evaluation interval Qr, an r-level warning signal is generated, the print collection accuracy is evaluated to be r-level, and it is output to the data display unit for visual presentation; when the print collection accuracy coefficient Xyh is lower, the level of the warning signal is lower, which means that the evaluation print collection accuracy is lower;

[0091] Sb-3, then use the depth analysis module to perform texture feature analysis on the print image, obtain the skin abnormality coefficient and development abnormality coefficient, and evaluate the skin surface abnormality and bone development abnormality of the print. The specific process of texture feature analysis is as follows:

[0092] Sb-301, first construct the gray level co-occurrence matrix, extract N0 groups of neighboring pixel pairs, mark any group of neighboring pixel pairs as G, take the horizontal direction as the reference direction, the neighboring pixel pair G includes pixel g1 and pixel g2, compare the gray value of any pixel g1 with its adjacent pixel g2, and measure the gray difference Δg of the neighboring pixel pair G: Δg = |g1-g2|;

[0093] Sb-302, classifying the N0 groups of neighboring pixel pairs according to the grayscale difference Δg, accumulating the frequency Jg of the neighboring pixel pairs with the grayscale difference Δg in the N0 groups of neighboring pixel pairs, and then obtaining the grayscale difference Δgm corresponding to the highest frequency Jg by sorting in descending order;

[0094] Sb-303, preset a grayscale difference threshold φg. When the grayscale difference Δg is higher than the grayscale difference threshold φg, mark the group of neighboring pixel pairs as YCg, extract the pixel points with high grayscale value in the neighboring pixel pairs YCg, and mark them with characteristic colors. Thus, through the comparative analysis of N0 groups of neighboring pixel pairs, obtain all the pixel points Co marked with characteristic colors, and mark the closed area formed by connecting the pixel points Co as the characteristic texture area.

[0095] Sb-303-1, perform grayscale co-occurrence matrix analysis on the characteristic texture area, obtain the grayscale difference Δgz corresponding to the characteristic texture area, set the floating interval Qd[Δgm-z, Δgm+z] of the grayscale difference Δgz, when the grayscale difference Δgz exceeds the floating interval Qd, the characteristic texture area is determined to be a skin abnormality area, and there may be skin abnormalities such as herpes and scratches;

[0096] Then, the number and area of ​​abnormal skin regions are measured and marked as Np and Sp respectively, and then the skin abnormality coefficient Xpf is obtained by combining the number Np and area Sp of abnormal skin regions: Xpf = α1*Np+α2*Sp;

[0097] Among them, α1 and α2 are weight factor coefficients of the number Np and area Sp of skin abnormality areas, respectively, and α1 and α2 are both greater than 0, and the weight factor coefficients are obtained by preset; when the number Np and area Sp of skin abnormality areas are higher, the skin abnormality coefficient Xpf is higher, indicating that the abnormality of the foot skin is more serious;

[0098] Sb-303-2, partition the foot structure of the impression image P1 according to professional medical knowledge, and obtain the arch area and the toe area: if the characteristic texture area is located in the arch area, it is marked as the arch area Z0 and an arch abnormality analysis is performed; if the characteristic texture area is located in the toe area, it is marked as the toe area Z1 and an toe abnormality analysis is performed;

[0099] Sb-303-201, the specific process of arch abnormality analysis is:

[0100] The standard proportion of the arch area ZG is set, and the standard arch area Sa0 of the imprint image P1 is calculated based on the footprint area of ​​the imprint image P1, and the actual area Sa1 of the arch area Z0 is measured to obtain the arch area deviation Pg: When the arch area deviation Pg is higher, it means that there may be arch deformation, indicating that the newborn has abnormal motor development;

[0101] Sb-303-202, the specific process of toe abnormality analysis is:

[0102] The standard proportion ZB of the toe area is set, and the standard toe area Sb0 of the print image P1 is calculated according to the footprint area of ​​the print image P1, and the actual area Sb1 of the toe area Z1 is measured to obtain the toe area deviation Pz: The higher the toe area deviation Pz is, the more likely it is that there is toe abnormality, indicating that the newborn has abnormal toe morphology such as too large toe gap or too small toe bones.

[0103] Sb-303-3, and then obtain the developmental abnormality coefficient Xfy by combining the arch area deviation Pg and the toe area deviation Pz: Xfy = βα1*Pg+β2*Pz;

[0104] Among them, β1 and β2 are weight factor coefficients of arch area deviation Pg and toe area deviation Pz, respectively, and β1 and β2 are both greater than 0, and the weight factor coefficients are obtained by preset; when the arch area deviation Pg and toe area deviation Pz are higher, the developmental abnormality coefficient Xfy is higher, indicating that the abnormal development of foot bones is more serious;

[0105] Sb-4, through the combination of skin abnormality coefficient and developmental abnormality coefficient, comprehensively obtains the health assessment index and evaluates the physical health status of the newborn, and generates prompt signals and footprint health reports for visual presentation. The specific process of evaluating the physical health status of the newborn is as follows:

[0106] Sb-401, by combining the skin abnormality coefficient Xpf and the developmental abnormality coefficient Xfy, comprehensively obtains the health assessment index JK:

[0107] Among them, ε1 and ε2 are weight factor coefficients of skin abnormality coefficient Xpf and developmental abnormality coefficient Xfy, respectively, and both ε1 and ε2 are greater than 0; when the skin abnormality coefficient Xpf and developmental abnormality coefficient Xfy are higher, the health assessment index JK is lower, indicating that the health status of the foot signs of the newborn is lower;

[0108] Sb-402, set the standard interval of the health assessment index JK, evaluate the health of the newborn's physical signs by comparing the intervals, and generate prompt signals and footprint health reports for visual presentation:

[0109] Sb-402-1, there are U standard intervals preset, and any standard interval is marked as Qu, where u is the serial number of the standard interval, 0<u≤U, when the health assessment index JK is in the standard interval Qu, a u-level prompt signal is generated, and the physical health level of the newborn is assessed as u-level, and it is output to the data display unit for visual presentation; when the health assessment index JK is lower, the level of the prompt signal is lower, which means that the physical health level of the newborn is assessed to be lower;

[0110] Sb-402-2, and then integrate the skin abnormality coefficient Xpf, developmental abnormality coefficient Xfy and health assessment index JK to generate a footprint health report, and output it to the data display unit for visualization for reference by professional medical staff. Finally, through intelligent analysis and visualization, the accuracy and efficiency of newborn footprint analysis are continuously improved;

[0111] Sc, the data display unit receives the signal and the health report of the footprint trace, and makes corresponding visual presentation for reference by professional medical staff.

[0112] In summary, the present invention collects the initial print image through the print collection unit, and then performs print analysis and processing through the image processing unit, evaluates the print collection accuracy and analyzes the health status of the newborn's physical signs, and then performs early warning prompts and visual presentation through the data display unit, thereby continuously improving the print collection accuracy and ensuring the stability of the print collector. It is suitable for archiving and registering the footprints of newborns after pregnancy, and collecting and analyzing the footprints through regular physical examinations of newborns, which can assist in medical diagnosis and health assessment of infants and young children, and understand the growth and development status and possible abnormalities.

[0113] Among them, the present invention enhances the contrast of the initial print image through grayscale enhancement processing, and obtains a refined print image for subsequent texture feature analysis. The skin abnormalities on the surface of the footprint and the bone abnormalities of the toes and arches of the footprint are analyzed through the print texture features, so as to comprehensively evaluate the health status of the newborn represented by the footprint traces, so as to provide early warning of potential growth and development health problems of the newborn. The same can be done for hand prints.

[0114] The footprints and handprints produced during the above process can also be printed into a commemorative album or a commemorative electronic album.

[0115] The print collection unit collects relevant information through a high-definition camera collector, or can also form a preliminary print image by sensing the pressure information of the foot or hand formed by the film sensor after squeezing the sensing end.

[0116] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by technical personnel in this field for each group of sample data; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0117] The above formulas are all dimensionless and numerical calculations. The formula is a formula obtained by collecting a large amount of data and performing software simulation to obtain the most recent real situation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0118] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A print collection visualization system, characterized in that: It includes an imprint acquisition unit, an image processing unit and a data display unit, and the imprint acquisition unit, the image processing unit and the data display unit are connected by signals, wherein the image processing unit includes a preprocessing module and a depth analysis module, and the preprocessing module and the depth analysis module are connected by signals; The impression collection unit is used to collect at least one initial impression image and store it in the maternal electronic medical record, and then upload the initial impression image to the image processing unit; The image processing unit is used to analyze and process the initial printed image: The specific process of impression analysis is as follows: The grayscale enhancement process is performed on the initial printing image through the pre-processing module to obtain the print image; By comparing the initial print image with the print image, the print acquisition accuracy coefficient is obtained, so as to analyze and evaluate the print acquisition accuracy and generate corresponding warning signals; Then, the texture feature analysis of the print image is performed through the depth analysis module to obtain the skin abnormality coefficient and the developmental abnormality coefficient, and to evaluate the skin surface abnormality and bone developmental abnormality of the print; By combining the skin abnormality coefficient and the developmental abnormality coefficient, a comprehensive health assessment index is obtained to evaluate the physical health of the newborn, and a prompt signal and footprint health report are generated for visual presentation; The grayscale enhancement processing of the initial print image is performed through the preprocessing module to generate an imprint image, and the initial print image and the imprint image are compared to analyze and evaluate the imprint acquisition accuracy and generate an early warning signal; The specific process of grayscale enhancement processing is: The first printed image is marked as P0, and the image P0 is preset to have n0 feature points. Any feature point is marked as a. A regular quadrilateral grid area is delineated with the feature point a as the center, and the side length of the regular quadrilateral grid is preset to be d, so as to obtain and mark it as a grid Wa; There are n1 points in the preset grid Wa. The grayscale values ​​of all points in the grid Wa are acquired through image processing software, and any point in the grid Wa is marked as b, and the grayscale value of point b is marked as Hb. A discrete distribution graph is established for data display. First, obtain the median ZWa of the grayscale values ​​of n1 points in the grid Wa, and obtain the average PJa of the grayscale values ​​of n1 points; Through the median ZWa and average PJa of the grayscale value, a formula is established to obtain the grayscale conversion standard value Ha; Then, the standard deviation σb of the gray value is obtained, and the kurtosis coefficient FDb of the gray value and the skewness coefficient PDb of the gray value are calculated; The grayscale difference contrast value Zcy is obtained by combining the standard deviation σb, the kurtosis coefficient FDb and the skewness coefficient PDb of the grayscale value; The threshold F0 of the grayscale difference contrast value Zcy is set, and then the threshold comparison is performed to convert and replace the grayscale values ​​of all points of the grid Wa. After the grayscale data of the grid area where the n0 feature points of the image P0 are located is converted and adjusted in sequence, the imprint image is generated and marked as P1; The specific process of transformation and replacement is: If the grayscale difference contrast value Zcy is less than the threshold F0, the grid Wa is transformed as a whole, and the grayscale values ​​of all points of the grid Wa are adjusted to the grayscale transformation standard value Ha; If the grayscale difference comparison value Zcy ≥ threshold F0, the grayscale values ​​of all points of the grid Wa are compared with the grayscale conversion standard value Ha in turn: if the grayscale value of point b is lower than or equal to the grayscale conversion standard value Ha, no processing is performed; if the grayscale value of point b is higher than the grayscale conversion standard value Ha, the grayscale value of point b is enhanced, specifically: set the grayscale enhancement range φ%, obtain the grayscale enhancement value Hd through the grayscale enhancement range φ%, and adjust the grayscale value of point b to Hd; The specific process of analyzing and evaluating the accuracy of impression collection is as follows: Place the initial print image P0 and the imprint image P1 in the same coordinate system, extract m1 points of the initial print image P0, mark any point as e, and mark the point of the imprint image P1 with the same coordinate as the point as f; The gray value of point e is marked as He, and the gray value of point f is marked as Hf, and then the gray difference coefficient Xh is obtained; Then, the grayscale difference coefficient of point m1 is used to comprehensively obtain the print collection accuracy coefficient Xyh; Set the evaluation interval of the impression collection accuracy coefficient Xyh, evaluate the impression collection accuracy through interval comparison and generate corresponding warning signals: There are R evaluation intervals preset, and any evaluation interval is marked as Qr, where r is the serial number of the evaluation interval, 0<r≤R, and when the print collection accuracy coefficient Xyh is in the evaluation interval Qr, an r-level warning signal is generated, and the print collection accuracy is evaluated as r-level, and it is output to the data display unit for visual presentation; The deep analysis module then analyzes the texture features of the footprint image to assess the abnormalities of the skin surface and bone development of the footprint, and then comprehensively analyzes and assesses the physical health of the newborn, and generates prompt signals and footprint health reports; The specific process of texture feature analysis is: First, construct a gray-level co-occurrence matrix, extract N0 groups of adjacent pixel pairs, mark any group of adjacent pixel pairs as G, take the horizontal direction as the reference direction, the adjacent pixel pair G includes pixel g1 and pixel g2, compare the gray value of any pixel g1 with its adjacent pixel g2, and measure the gray difference of the adjacent pixel pair G g; According to grayscale difference g classifies N0 groups of adjacent pixel pairs, and accumulates the grayscale differences among N0 groups of adjacent pixel pairs. The frequency Jg of the neighboring pixel pairs of g is then sorted in descending order to obtain the grayscale difference corresponding to the highest frequency Jg. gm; Preset grayscale difference threshold φg, when grayscale difference When g is higher than the grayscale difference threshold φg, the group of neighboring pixel pairs is marked as YCg, and the pixels with high grayscale values ​​in the neighboring pixel pairs YCg are extracted and marked with characteristic colors. Then, through the comparative analysis of the N0 group of neighboring pixel pairs, all the pixels Co marked with characteristic colors are obtained, and the closed area formed by connecting the pixel points Co is marked as the characteristic texture area; Perform grayscale co-occurrence matrix analysis on the characteristic texture area to obtain the grayscale difference corresponding to the characteristic texture area gz, set grayscale difference The floating interval of gz is Qd[ gm-z, gm+z], when the grayscale difference When gz exceeds the floating interval Qd, the characteristic texture area is determined to be an abnormal skin area; Then, the number and area of ​​abnormal skin regions are measured and marked as Np and Sp respectively, and then the skin abnormality coefficient Xpf is obtained by combining the number Np and area Sp of abnormal skin regions; According to professional medical knowledge, the print image P1 is partitioned into foot structures to obtain the arch area and the toe area: if the characteristic texture area is located in the arch area, it is marked as the arch area Z0 and an arch abnormality analysis is performed; if the characteristic texture area is located in the toe area, it is marked as the toe area Z1 and an toe abnormality analysis is performed; The arch area deviation degree Pg is obtained by analyzing the abnormality of the arch, the toe area deviation degree Pz is obtained by analyzing the abnormality of the toe, and then the developmental abnormality coefficient Xfy is obtained by combining the arch area deviation degree Pg and the toe area deviation degree Pz; The data display unit is used to receive signals and footprint health reports, and make corresponding visual presentations for reference by professional medical staff.

2. The impression collection and visualization system according to claim 1, characterized in that: The specific process of arch abnormality analysis and toe abnormality analysis is as follows: Arch abnormality analysis: Set the arch area standard proportion ZG, calculate the arch standard area Sa0 of the imprint image P1 according to the footprint area of ​​the imprint image P1, and measure the actual area Sa1 of the arch area Z0 to obtain the arch area deviation Pg; Toe abnormality analysis: Set the toe area standard proportion ZB, calculate the toe standard area Sb0 of the print image P1 based on the footprint area of ​​the print image P1, and measure the actual area Sb1 of the toe area Z1 to obtain the toe area deviation Pz.

3. The impression collection and visualization system according to claim 2, characterized in that: The specific process of assessing the physical health status of a newborn is as follows: By combining the skin abnormality coefficient Xpf and the developmental abnormality coefficient Xfy, the health assessment index JK is comprehensively obtained; Set the standard interval of the health assessment index JK, evaluate the health of the newborn's physical signs through interval comparison, and generate prompt signals and footprint health reports for visual presentation: There are U standard intervals preset, and any standard interval is marked as Qu, where u is the serial number of the standard interval, 0<u≤U. When the health assessment index JK is in the standard interval Qu, a u-level prompt signal is generated, and the physical health level of the newborn is assessed as u-level, and it is output to the data display unit for visual presentation; The skin abnormality coefficient Xpf, developmental abnormality coefficient Xfy and health assessment index JK are then integrated to generate a footprint health report, which is then output to a data display unit for visual presentation for reference by professional medical staff.

4. The impression collection and visualization system according to claim 1, characterized in that: The initial print is a footprint image or a handprint image.

5. An impression collector, characterized in that: The invention comprises the impression collection and visualization system as described in any one of claims 1 to 4.

Images