An image processing and printing system through camera sampling

Abstract

The present invention discloses an image processing and printing system through camera sampling, comprising a data acquisition module, a pre-storage module, a pre-processing module, a data processing module and a printing module; the pre-processing module is used for screening target image data and image feature data of a target person, and establishing a preset local model, and then trimming the preset local model to obtain printing model data; the data processing module is used for obtaining printing image data; the present invention continuously obtains the image data of the target person through a specific method, and controls the shooting angle and number of images through analysis of the image data, thereby improving the validity of the image data; then the image data is analyzed to perform character modeling, and finally the character modeling is adjusted according to the feature data of the target person, so as to retain the real features of the face of the target person to a greater extent, reduce the probability of photo distortion, and improve the quality of equipment shooting.

Description

Technical Field

[0001] The present invention relates to the field of intelligent printing, and in particular to an image processing and printing system through camera sampling. Background Art

[0002] As people's living standards improve, the demand for photos is also increasing. The all-in-one camera and printer can quickly capture and efficiently produce photos, meeting people's needs for recording and sharing beautiful moments. This device is particularly suitable for family gatherings, friends' dinners, travel souvenirs and other occasions, providing people with convenient and fast printing services; however, the existing printing equipment has a low level of intelligence, generally collecting people's pictures through the camera and transmitting them to the printing module for printing; because the camera environment and shooting method and other factors have a great impact on the quality of the photos, the quality of the photos cannot be controlled, resulting in serious distortion of the printed pictures and poor appearance; at the same time, the shooting method of the photos is fixed, and it takes multiple times to shoot pictures that the user is satisfied with, which has a poor user experience and affects the marketing and development of the device; to solve this problem, an image processing and printing system through camera sampling is proposed. Summary of the invention

[0003] The technical problem to be solved by the present invention is: how to solve the problem that the existing security protection system has a single protection type, resulting in poor protection effect and bringing certain impacts on the use of the security protection system. An image processing and printing system is provided through camera sampling.

[0004] The present invention solves the above technical problems through the following technical solutions. The present invention includes a data acquisition module, a pre-storage module, a pre-processing module, a data processing module, and a printing module;

[0005] The data acquisition module is used to collect real-time data information of the target person, and the real-time data information includes real-time image data;

[0006] The pre-storage module is used to store the real-time image data of the target person within a preset period of time;

[0007] The preprocessing module is used to receive real-time image data, screen out multiple target image data and image feature data of a target person through processing and analysis of the real-time image data, establish a preset local model of the target person based on the multiple target image data, and then trim the preset local model through the image feature data to obtain the print model data of the target person;

[0008] The data processing module is used to receive the printing model data, adjust the printing position of the printing model data, and obtain the printing image data;

[0009] The printing module is used to identify the printing image data and print the picture according to the printing image data.

[0010] Preferably, the real-time data information also includes real-time size data, and the specific processing process of the image acquisition module is:

[0011] Randomly collect real-time image data of the target person;

[0012] Obtain the intersection line between the philtrum and the nose of the target person, and mark the midpoint of the intersection line as point A1;

[0013] Then obtain the eye positions of the target person, and make second marks at the geometric centers of the two eyes, which are recorded as points A2 and A3;

[0014] Connect point A1, point A2 and point A3 in sequence to establish a virtual triangle;

[0015] Get the length L1 of the virtual triangle side A1A2 and the length L2 of the virtual triangle side A1A3;

[0016] Calculate the difference E1 between L1 and L2. The specific calculation process is:

[0017] ;

[0018] When the difference E1 is less than the preset threshold Q1, the real-time image data is defined as the first type of image data;

[0019] Taking the first type of image data as a reference position, a plurality of real-time image data of the target person are continuously collected on both sides around the facial features of the target person in the horizontal direction;

[0020] Randomly extract a real-time image data B1, and obtain the facial contour line C1 and the head contour line C2 of the target person in the real-time image data B1;

[0021] Segment the head contour line C2 into a plurality of unit rectangles;

[0022] Count the data Da1 of the complete unit rectangles and the number Da2 of the incomplete unit rectangles within the facial contour line C1;

[0023] Then count the data Db1 of the complete unit rectangles and the number Db2 of the incomplete unit rectangles within the head contour line C2;

[0024] Calculate the facial contour ratio E2. The specific calculation process is:

[0025] ;

[0026] When the facial contour proportion E2 is less than the preset threshold Q2, the corresponding real-time image data is cleared.

[0027] Preferably, the process of acquiring the facial contour line C1 further includes:

[0028] Import real-time image data;

[0029] Use preset rules to select points on one side of the target person's face to create a local virtual contour line;

[0030] The default rules are:

[0031] Obtain the eyebrow position of the target person, and make third marks at the geometric midpoints of the two eyebrows, recorded as point A4 and point A5;

[0032] Connect point A4 and point A5 to create a first virtual straight line La;

[0033] A second virtual straight line Lb is made on the forehead of the target person, which is parallel to the first virtual straight line La. The distance between the second virtual straight line Lb and the first virtual straight line La is equal to the distance between the first virtual straight line La and the base of the nose of the target person.

[0034] Obtain the outer corner of the eye and the ear of the target person, and make a fourth mark at the outer corner of the eye, recorded as point A6;

[0035] Take the lowest point of the ear for the fifth mark, record it as point A7;

[0036] Obtain the midpoint of the line segment where point A6 and point A7 are located for the sixth marking, recorded as A8;

[0037] A third virtual straight line Lc is drawn through point A8 and is parallel to the midline of the face. The intersection of the second virtual straight line Lb and the third virtual straight line Lc is the first reference point As1. The intersection of the first virtual straight line La and the third virtual straight line Lc is the second reference point As2. The geometric center of the ear is the third reference point As3.

[0038] Connect the first reference point As1, the second reference point As2 and the third reference point As3 to form a local virtual contour line;

[0039] Using a preset rule, points are taken on both sides of the target person's face to create a first local virtual contour line C3 and a second local virtual contour line C4, and the first local virtual contour line C3 and the second local virtual contour line C4 are connected to form a local virtual contour line C5 of the face;

[0040] Obtain the facial contour line C1 of the target person, and use the local virtual facial contour C5 to replace the corresponding data of the facial contour line C1 to form a virtual facial contour line C6;

[0041] Count the data Dc1 of the complete unit rectangles and the number Dc2 of the incomplete unit rectangles within the virtual facial contour line C6;

[0042] The first deviation E3 between the virtual facial contour line C6 and the facial contour line C1 is calculated. The specific calculation process is:

[0043] ;

[0044] When the first deviation degree E3>the preset threshold value Q3, the facial contour line C1 is changed into the virtual facial contour line C6.

[0045] Preferably, the target image data is generated by:

[0046] Import multiple real-time image data of one side of the face and obtain multiple facial contour lines, namely Cr1, Cr1, Cr1, and Crn;

[0047] Obtain data De1 of complete unit rectangles and the number De2 of incomplete unit rectangles in the facial contour line of the nth real-time image data;

[0048] Then import the first type of image data to obtain the facial contour line Ci of the first type of image data;

[0049] Obtain data Di1 of complete unit rectangles and the number Di2 of incomplete unit rectangles in the facial contour line Ci of the first type image data;

[0050] The second deviation E4 between the nth real-time image data and the first type of image data is calculated. The specific calculation process is:

[0051] ;

[0052] Calculate the second deviation E4 of each real-time image data and the first type of image data respectively, and sort them in order;

[0053] Obtaining second deviations of two adjacent real-time image data, which are E4a and E4b respectively;

[0054] Calculate the similarity E5 of two adjacent real-time images respectively. The specific calculation process is:

[0055] ;

[0056] When the familiarity E5>preset threshold Q4, one of the two real-time image data is cleared to generate target image data.

[0057] Preferably, the process of generating the target image data further includes:

[0058] Acquire the first type of image data, and randomly acquire a detection point T at a preset distance from the face center line of the target person's face;

[0059] Read the target image data of the side surface corresponding to the detection point T;

[0060] Count the number of target image data Z1;

[0061] Count the number Z2 of target image data with detection point T;

[0062] Use the formula E6=Z2 / Z1 to calculate the shooting proportion E6 of the detection point T;

[0063] When the shooting proportion E6 is less than the preset threshold Q5, the preset threshold Q4 is adjusted to Q4`, and Q4` is greater than Q4.

[0064] Preferably, the generation process of the printing model data is:

[0065] Importing first type image data, and taking the first type image data as reference data;

[0066] Then import the target image data on both sides of the target person's face;

[0067] According to the order of the real-time image data, based on the benchmark data, the benchmark data and the target image data are sequentially merged in order of order to generate a preset local model;

[0068] Divide the preset local model into regions;

[0069] Acquire multiple image information of each area respectively according to multiple target image data;

[0070] Marking and filtering image feature data in multiple image information of each area;

[0071] Obtaining corresponding image feature data in a preset local model;

[0072] Compare the image feature data in the preset local model with the image feature data in the regional image to obtain difference data;

[0073] When there is data missing in the preset partial model, the difference data is supplemented in the preset partial model to generate the printing model data of the target person.

[0074] Preferably, the specific processing process of printing image data is:

[0075] The printing model data of the target person is obtained, the printing model data is adjusted to the position of the first type of image data, and the first type of image data is obtained to generate the printing image data.

[0076] Preferably, the system further comprises a database module and a service platform module, wherein the database module is used to store the printing model data and printing image data of the target person, and the service platform module is used to receive and display the printing model data and adjust the printing position to generate the preferred image printing data. The processing process of the data processing module further comprises:

[0077] Acquire the printing model data, adjust the printing model data to the position of the first type of image data, make two mutually perpendicular virtual planes through the midline of the face, namely virtual plane M1 and virtual plane M2, and virtual plane M1 is the median vertical plane of the printing model data;

[0078] Randomly acquire a first monitoring point G1 and a second monitoring point G2 on the virtual plane M2, wherein the first monitoring point G1 and the second monitoring point G2 are symmetrically arranged with respect to the virtual plane M1;

[0079] Draw a first perpendicular line through the first monitoring point G1 to the virtual plane M2, and the intersection of the first perpendicular line and the printed model is point G3;

[0080] A second perpendicular line is drawn through the second monitoring point G2 to the virtual plane M2, and the intersection of the second perpendicular line and the printed model is point G4;

[0081] Collect the dimension data Lg1 from the first monitoring point G1 to the point G3, and collect the dimension data Lg2 from the second monitoring point G2 to the point G4;

[0082] Calculate the difference K between the size data Lg1 and the size data Lg2. The specific calculation process is:

[0083] ;

[0084] When the preset threshold Q6 < the difference degree K < the preset threshold Q7, the average value of the size data Lg1 and the size data Lg2 is obtained. ;

[0085] Adjust the size data Lg1 and size data Lg2 to To generate preferred printing model data and send the preferred printing model data to the service platform module.

[0086] Preferably, the process of generating the preferred printing model data further includes:

[0087] Import printing model data;

[0088] Obtaining the location area of ​​facial features in the printing model data;

[0089] When any one of the points G3 or G4 is located in the facial feature position area in the print model data, the size data Lg1 and the size data Lg2 are retained.

[0090] Compared with the prior art, the present invention has the following advantages: the system continuously acquires the image data of the target person in a specific manner, controls the shooting angle and number of images by analyzing the image data, removes pictures that do not meet the requirements, and improves the validity of the image data; then analyzes the image data to screen out a certain number of pictures for character modeling, and adjusts the character modeling according to the characteristic data of the target person to retain the real features of the target person's face to a greater extent, reduce the probability of photo distortion, and improve the quality of equipment shooting; at the same time, the system can collect the size data of the target person's face, fine-tune abnormal parts, optimize the facial details of the target person, and display the optimized character modeling in the service platform module for consumers to choose the printing position by themselves, so as to select picture data that satisfies the consumers, avoid the tedious process of multiple operations by users, improve the user experience, and also improve the market competitiveness of the system, making the system more worthy of promotion and use. BRIEF DESCRIPTION OF THE DRAWINGS

[0091] Figure 1 It is the overall module diagram of the present invention. DETAILED DESCRIPTION

[0092] The following is a detailed description of an embodiment of the present invention. This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and a specific operation process are given, but the protection scope of the present invention is not limited to the following embodiment.

[0093] like Figure 1 As shown, this embodiment provides a technical solution: an image processing and printing system through camera sampling, including a data acquisition module, a pre-storage module, a pre-processing module, a data processing module, and a printing module;

[0094] The data acquisition module is used to collect real-time data information of the target person, and the real-time data information includes real-time image data;

[0095] The pre-storage module is used to store the real-time image data of the target person within a preset period of time. The pre-storage module is used to temporarily store the image data of the consumer and provide data support for system analysis within the preset period of time; to avoid data occupying a large amount of system memory and affecting its operating efficiency.

[0096] The preprocessing module is used to receive real-time image data, screen out multiple target image data and image feature data of a target person through processing and analysis of the real-time image data, establish a preset local model of the target person based on the multiple target image data, and then trim the preset local model through the image feature data to obtain the print model data of the target person;

[0097] It should be noted that the image features include facial scratches, special colors, moles, etc. of the target person, and also include facial features of the target person, such as eyes, eyebrows, nose, mouth, ears and other facial feature data.

[0098] The data processing module is used to receive the printing model data, adjust the printing position of the printing model data, and obtain the printing image data;

[0099] The printing module is used to identify the printing image data and print the picture according to the printing image data.

[0100] The system continuously acquires the image data of the target person in a specific way, and controls the shooting angle and number of images by analyzing the image data, thereby removing pictures that do not meet the requirements and improving the validity of the image data. The system then analyzes the image data to select a certain number of pictures for character modeling, and adjusts the character modeling according to the characteristic data of the target person to retain the true features of the target person's face to a greater extent, reduce the probability of photo distortion, and improve the quality of equipment shooting.

[0101] Among them, the real-time data information also includes real-time size data. The specific processing process of the image acquisition module is:

[0102] Randomly collect real-time image data of the target person;

[0103] Obtain the intersection line between the philtrum and the nose of the target person, and mark the midpoint of the intersection line as point A1;

[0104] Then obtain the eye positions of the target person, and make second marks at the geometric centers of the two eyes, which are recorded as points A2 and A3;

[0105] Connect point A1, point A2 and point A3 in sequence to establish a virtual triangle;

[0106] Get the length L1 of the virtual triangle side A1A2 and the length L2 of the virtual triangle side A1A3;

[0107] Calculate the difference E1 between L1 and L2. The specific calculation process is:

[0108] ;

[0109] When the difference E1 is less than the preset threshold Q1, the real-time image data is defined as the first type of image data;

[0110] Taking the first type of image data as a reference position, a plurality of real-time image data of the target person are continuously collected on both sides around the facial features of the target person in the horizontal direction;

[0111] First, the image data of the target person's front face is obtained, and then the image data of the target person is collected from both sides along the front face of the target person.

[0112] Randomly extract a real-time image data B1, and obtain the facial contour line C1 and the head contour line C2 of the target person in the real-time image data B1;

[0113] Segment the head contour line C2 into a plurality of unit rectangles;

[0114] It should be noted that the unit rectangle is a rectangle made according to a preset basic unit.

[0115] Count the data Da1 of the complete unit rectangles and the number Da2 of the incomplete unit rectangles within the facial contour line C1;

[0116] Then count the data Db1 of the complete unit rectangles and the number Db2 of the incomplete unit rectangles within the head contour line C2;

[0117] Calculate the facial contour ratio E2. The specific calculation process is:

[0118] ;

[0119] When the facial contour proportion E2 is less than the preset threshold Q2, the corresponding real-time image data is cleared.

[0120] This embodiment first obtains the front face data of the target person, that is, the first type of image data, by analyzing the graphic data of the target person's face; then collects the real-time image data of the target person on both sides based on the first type of image data; then divides the real-time image data, replaces the area with the number of unit rectangles, and calculates the ratio of the real-time image facial contour area to the entire head contour area; when the ratio decreases, it means that the real-time image data is the side image data of the target person far from the front, and the data is not within the scope of the consumer's printed picture, so it is deleted; through this method, the acquisition range and quantity of the image are controlled, and then the shooting angle of the image is controlled, the effectiveness and reference of the real-time data are improved, and the system data analysis is more targeted.

[0121] In the process of real-time image data acquisition, when the target person wears something on his head or has bangs, it will affect the accuracy of obtaining the target person's facial contour line. To solve this problem, the following further solutions are proposed:

[0122] Furthermore, the process of acquiring the facial contour line C1 also includes:

[0123] Import real-time image data;

[0124] Use preset rules to select points on one side of the target person's face to create a local virtual contour line;

[0125] The default rules are:

[0126] Obtain the eyebrow position of the target person, and make third marks at the geometric midpoints of the two eyebrows, recorded as point A4 and point A5;

[0127] Connect point A4 and point A5 to create a first virtual straight line La;

[0128] A second virtual straight line Lb is made on the forehead of the target person, which is parallel to the first virtual straight line La. The distance between the second virtual straight line Lb and the first virtual straight line La is equal to the distance between the first virtual straight line La and the base of the nose of the target person.

[0129] Obtain the outer corner of the eye and the ear of the target person, and make a fourth mark at the outer corner of the eye, recorded as point A6;

[0130] Take the lowest point of the ear for the fifth mark, record it as point A7;

[0131] Obtain the midpoint of the line segment where point A6 and point A7 are located for the sixth marking, recorded as A8;

[0132] A third virtual straight line Lc is drawn through point A8 and is parallel to the midline of the face. The intersection of the second virtual straight line Lb and the third virtual straight line Lc is the first reference point As1. The intersection of the first virtual straight line La and the third virtual straight line Lc is the second reference point As2. The geometric center of the ear is the third reference point As3.

[0133] It should be noted that the facial midline is the longitudinal center line of the target person's face.

[0134] Connect the first reference point As1, the second reference point As2 and the third reference point As3 to form a local virtual contour line;

[0135] Using a preset rule, points are taken on both sides of the target person's face to create a first local virtual contour line C3 and a second local virtual contour line C4, and the first local virtual contour line C3 and the second local virtual contour line C4 are connected to form a local virtual contour line C5 of the face;

[0136] Obtain the facial contour line C1 of the target person, and use the local virtual facial contour C5 to replace the corresponding data of the facial contour line C1 to form a virtual facial contour line C6;

[0137] Count the data Dc1 of the complete unit rectangles and the number Dc2 of the incomplete unit rectangles within the virtual facial contour line C6;

[0138] The first deviation E3 between the virtual facial contour line C6 and the facial contour line C1 is calculated. The specific calculation process is:

[0139] ;

[0140] When the first deviation degree E3>the preset threshold value Q3, the facial contour line C1 is changed into the virtual facial contour line C6.

[0141] In this scheme, a preset rule is first used to obtain a local virtual facial contour of the target person, and the virtual contour is inferred based on the target person's face shape; when the target person's face is occluded, the virtual facial contour where the local virtual facial contour is located is calculated and compared with the facial contour obtained by the system. When the difference between the two is small, it means that the occluded part of the target person's face is not enough to affect the collection of the facial contour line and can be ignored; when the difference between the two is large, it means that the occluded part of the target person's face is enough to affect the accuracy of the facial contour line collection, and the virtual facial contour is used to replace the facial contour obtained by the system to reduce the calculation error of the system and improve the accuracy and reliability of the system.

[0142] Among them, the generation process of target image data is:

[0143] Import multiple real-time image data of one side of the face and obtain multiple facial contour lines, namely Cr1, Cr1, Cr1, and Crn;

[0144] Obtain data De1 of complete unit rectangles and the number De2 of incomplete unit rectangles in the facial contour line of the nth real-time image data;

[0145] Then import the first type of image data to obtain the facial contour line Ci of the first type of image data;

[0146] Obtain the data Di1 of the complete unit rectangle and the number Di2 of the incomplete unit rectangle in the facial contour line Ci of the first type image data

[0147] The second deviation E4 between the nth real-time image data and the first type of image data is calculated. The specific calculation process is:

[0148] ;

[0149] Calculate the second deviation E4 of each real-time image data and the first type of image data respectively, and sort them in order;

[0150] Obtaining second deviations of two adjacent real-time image data, which are E4a and E4b respectively;

[0151] Calculate the similarity E5 of two adjacent real-time images respectively. The specific calculation process is:

[0152] ;

[0153] When the familiarity E5>preset threshold Q4, one of the two real-time image data is cleared to generate target image data.

[0154] In this scheme, the degree of deviation between any real-time image data and the first type of image data is calculated; then the degree of similarity between two adjacent real-time image data is determined by comparing the degree of deviation between the two adjacent real-time image data. When the degree of similarity between two adjacent real-time image data is large, the referenceability of a single image data will be reduced; data with a large degree of similarity will be cleared to enhance the use value of the real-time image data and improve the fluency of system use.

[0155] In order to further screen out a reasonable amount of target image data and improve the accuracy of system data analysis, the following further solutions are proposed:

[0156] Furthermore, the process of generating the target image data also includes:

[0157] Acquire the first type of image data, and randomly acquire a detection point T at a preset distance from the face center line of the target person's face;

[0158] Read the target image data of the side surface corresponding to the detection point T;

[0159] Count the number of target image data Z1;

[0160] Count the number Z2 of target image data with detection point T;

[0161] Use the formula E6=Z2 / Z1 to calculate the shooting proportion E6 of the detection point T;

[0162] When the shooting proportion E6 is less than the preset threshold Q5, the preset threshold Q4 is adjusted to Q4`, and Q4` is greater than Q4.

[0163] Detection points are randomly acquired on the target person's face for detection, and the coverage of multiple real-time image data on the detection points is calculated; when the coverage is small, it indicates that the amount of real-time image data cleared is too large, and the value of the preset threshold Q4 needs to be increased to reduce the number of real-time images cleared and increase the number of target image data samples to ensure the accuracy of system data analysis.

[0164] The generation process of printing model data is as follows:

[0165] Importing first type image data, and taking the first type image data as reference data;

[0166] Then import the target image data on both sides of the target person's face;

[0167] According to the order of the real-time image data, based on the benchmark data, the benchmark data and the target image data are sequentially merged in order of order to generate a preset local model;

[0168] Divide the preset local model into regions;

[0169] Acquire multiple image information of each area respectively according to multiple target image data;

[0170] Marking and filtering image feature data in multiple image information of each area;

[0171] Obtaining corresponding image feature data in a preset local model;

[0172] Compare the image feature data in the preset local model with the image feature data in the regional image to obtain difference data;

[0173] When there is data missing in the preset partial model, the difference data is supplemented in the preset partial model to generate the printing model data of the target person.

[0174] First, character modeling is performed using the target image data, and then the character modeling is adjusted based on the target character's feature data to retain the target character's true facial features to a greater extent, reduce the probability of photo distortion, and improve the quality of device photography.

[0175] The specific processing process of printing image data is as follows:

[0176] The printing model data of the target person is obtained, the printing model data is adjusted to the position of the first type of image data, and the first type of image data is obtained to generate the printing image data.

[0177] Some users like more realistic photos to record themselves, while some users like more perfect photos to please themselves and show themselves. In order to meet the needs of different users, the following solutions are proposed

[0178] It also includes a database module and a service platform module. The database module is used to store the printing model data and printing image data of the target person. The service platform module is used to receive and display the printing model data and adjust the printing position to generate the preferred image printing data. The processing process of the data processing module also includes:

[0179] Acquire the printing model data, adjust the printing model data to the position of the first type of image data, make two mutually perpendicular virtual planes through the midline of the face, namely virtual plane M1 and virtual plane M2, and virtual plane M1 is the median vertical plane of the printing model data;

[0180] It should be noted that the median vertical plane is a plane passing through the midline of the face, and the facial features on both sides of the person are arranged symmetrically about the median vertical plane.

[0181] Randomly acquire a first monitoring point G1 and a second monitoring point G2 on the virtual plane M2, wherein the first monitoring point G1 and the second monitoring point G2 are symmetrically arranged with respect to the virtual plane M1;

[0182] Draw a first perpendicular line through the first monitoring point G1 to the virtual plane M2, and the intersection of the first perpendicular line and the printed model is point G3;

[0183] A second perpendicular line is drawn through the second monitoring point G2 to the virtual plane M2, and the intersection of the second perpendicular line and the printed model is point G4;

[0184] Collect the dimension data Lg1 from the first monitoring point G1 to the point G3, and collect the dimension data Lg2 from the second monitoring point G2 to the point G4;

[0185] Calculate the difference K between the size data Lg1 and the size data Lg2. The specific calculation process is:

[0186] ;

[0187] When the preset threshold Q6 < the difference degree K < the preset threshold Q7, the average value of the size data Lg1 and the size data Lg2 is obtained. ;

[0188] Adjust the size data Lg1 and size data Lg2 to To generate preferred printing model data and send the preferred printing model data to the service platform module.

[0189] By establishing different virtual planes and taking one of the virtual planes as a reference plane, the degree of size difference of the unevenness of the symmetrical position of the target person's face is calculated; when the degree of difference is within a preset range, adjustments are made to optimize the facial details of the target person and enhance the beauty of the target person's photo; when the degree of difference is less than the preset range, the slight difference between the two sides of the face is not enough to affect the beauty of the target person's photo and is ignored; when the degree of difference is greater than the preset range, it indicates that the difference is caused by the target person's facial defects or normal facial deformations such as smiles, and adjustments cannot be made to avoid photo distortion; in this way, the abnormal parts are fine-tuned, the facial details of the target person are optimized, and the optimized character modeling is displayed in the service platform module for consumers to choose the printing position by themselves, so as to select picture data that satisfies consumers, avoid the tedious process of multiple operations by users, improve the user experience, and also improve the market competitiveness of the system.

[0190] Preferably, the process of generating the preferred printing model data further includes:

[0191] Import printing model data;

[0192] Obtaining the location area of ​​facial features in the printing model data;

[0193] When any one of the points G3 or G4 is located in the facial feature position area in the print model data, the size data Lg1 and the size data Lg2 are retained.

[0194] When the target person's expression changes, the normal deformation of the facial features will cause facial asymmetry. This situation can show the natural beauty of the target person and cannot be adjusted. Special regulations should be made to ensure the naturalness of the face after the photo is printed.

[0195] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0196] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0197] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. An image processing and printing system for sampling by a camera, the system is used to obtain and print a target person image, characterized in that: It includes data acquisition module, pre-storage module, pre-processing module, data processing module and printing module; The data acquisition module is used to collect real-time data information of the target person, and the real-time data information includes real-time image data; The pre-storage module is used to store the real-time image data of the target person within a preset period of time; The preprocessing module is used to receive real-time image data, screen out multiple target image data and image feature data of a target person through processing and analysis of the real-time image data, establish a preset local model of the target person based on the multiple target image data, and then trim the preset local model through the image feature data to obtain the print model data of the target person; The data processing module is used to receive the printing model data, adjust the printing position of the printing model data, and obtain the printing image data; The printing module is used to identify the printing image data and print the image according to the printing image data; The real-time data information also includes real-time size data. The specific processing process of the data acquisition module is as follows: Randomly collect real-time image data of the target person; Obtain the intersection line between the philtrum and the nose of the target person, and mark the midpoint of the intersection line as point A1; Then obtain the eye positions of the target person, and make second marks at the geometric centers of the two eyes, which are recorded as points A2 and A3; Connect point A1, point A2 and point A3 in sequence to establish a virtual triangle; Get the length L1 of the virtual triangle side A1A2 and the length L2 of the virtual triangle side A1A3; Calculate the difference E1 between L1 and L2. The specific calculation process is: ； When the difference E1 is less than the preset threshold Q1, the real-time image data is defined as the first type of image data; Taking the first type of image data as a reference position, a plurality of real-time image data of the target person are continuously collected on both sides around the facial features of the target person in the horizontal direction; Randomly extract a real-time image data B1, and obtain the facial contour line C1 and the head contour line C2 of the target person in the real-time image data B1; Segment the head contour line C2 into a plurality of unit rectangles; Count the number of complete unit rectangles Da1 and the number of incomplete unit rectangles Da2 within the facial contour line C1; Then count the number Db1 of complete unit rectangles and the number Db2 of incomplete unit rectangles within the head contour line C2; Calculate the facial contour ratio E2. The specific calculation process is: ； When the facial contour proportion E2 is less than the preset threshold Q2, the corresponding real-time image data is cleared.

2. The image processing and printing system according to claim 1, characterized in that: The process of acquiring the facial contour line C1 also includes: Import real-time image data; Use preset rules to select points on one side of the target person's face to create a local virtual contour line; The default rules are: Obtain the eyebrow position of the target person, and make third marks at the geometric midpoints of the two eyebrows, recorded as point A4 and point A5; Connect point A4 and point A5 to create a first virtual straight line La; A second virtual straight line Lb is made on the forehead of the target person, which is parallel to the first virtual straight line La. The distance between the second virtual straight line Lb and the first virtual straight line La is equal to the distance between the first virtual straight line La and the base of the nose of the target person. Obtain the outer corner of the eye and the ear of the target person, and make a fourth mark at the outer corner of the eye, recorded as point A6; Take the lowest point of the ear for the fifth mark, record it as point A7; Obtain the midpoint of the line segment where point A6 and point A7 are located for the sixth marking, recorded as A8; A third virtual straight line Lc is drawn through point A8 and is parallel to the midline of the face. The intersection of the second virtual straight line Lb and the third virtual straight line Lc is the first reference point As1. The intersection of the first virtual straight line La and the third virtual straight line Lc is the second reference point As2. The geometric center of the ear is the third reference point As3. Connect the first reference point As1, the second reference point As2 and the third reference point As3 to form a local virtual contour line; Using a preset rule, points are taken on both sides of the target person's face to create a first local virtual contour line C3 and a second local virtual contour line C4, and the first local virtual contour line C3 and the second local virtual contour line C4 are connected to form a local virtual contour line C5 of the face; Obtain the facial contour line C1 of the target person, and use the local virtual facial contour C5 to replace the corresponding data of the facial contour line C1 to form a virtual facial contour line C6; Count the number Dc1 of complete unit rectangles and the number Dc2 of incomplete unit rectangles within the virtual facial contour line C6; The first deviation E3 between the virtual facial contour line C6 and the facial contour line C1 is calculated. The specific calculation process is: ； When the first deviation degree E3>the preset threshold value Q3, the facial contour line C1 is changed into the virtual facial contour line C6.

3. The image processing and printing system according to claim 1, characterized in that: The generation process of the target image data is as follows: Import multiple real-time image data of one side of the face to obtain multiple facial contour lines; Obtain the number De1 of complete unit rectangles and the number De2 of incomplete unit rectangles in the facial contour line of the nth real-time image data; Then import the first type of image data to obtain the facial contour line Ci of the first type of image data; Obtain the number Di1 of complete unit rectangles and the number Di2 of incomplete unit rectangles in the facial contour line Ci of the first type of image data; The second deviation E4 between the nth real-time image data and the first type of image data is calculated. The specific calculation process is: ； Calculate the second deviation E4 of each real-time image data and the first type of image data respectively, and sort them in order; Obtaining second deviations of two adjacent real-time image data, which are E4a and E4b respectively; Calculate the similarity E5 of two adjacent real-time images respectively. The specific calculation process is: ； When the similarity E5>preset threshold Q4, one of the two real-time image data is cleared to generate target image data.

4. The image processing and printing system according to claim 3, characterized in that: The target image data generation process further includes: Acquire the first type of image data, and randomly acquire a detection point T at a preset distance from the face center line of the target person's face; Read the target image data of the side surface corresponding to the detection point T; Count the number of target image data Z1; Count the number Z2 of target image data with detection point T; Use the formula E6=Z2 / Z1 to calculate the shooting proportion E6 of the detection point T; When the shooting proportion E6 is less than the preset threshold Q5, the preset threshold Q4 is adjusted to Q4`, and Q4` is greater than Q4.

5. The image processing and printing system according to claim 3, characterized in that: The generation process of the printing model data is as follows: Importing first type image data, and taking the first type image data as reference data; Then import the target image data on both sides of the target person's face; According to the order of the real-time image data, based on the benchmark data, the benchmark data and the target image data are sequentially merged in order of order to generate a preset local model; Divide the preset local model into regions; Acquire multiple image information of each area respectively according to multiple target image data; Marking and filtering image feature data in multiple image information of each area; Obtaining corresponding image feature data in a preset local model; Compare the image feature data in the preset local model with the image feature data in the regional image to obtain difference data; When there is data missing in the preset partial model, the difference data is supplemented in the preset partial model to generate the printing model data of the target person.

6. The image processing and printing system according to claim 3, characterized in that: The specific processing process of printing image data is as follows: The printing model data of the target person is obtained, the printing model data is adjusted to the position of the first type of image data, and the first type of image data is obtained to generate the printing image data.

7. The image processing and printing system according to claim 2 or 3, characterized in that: The system further includes a database module and a service platform module, wherein the database module is used to store the printing model data and printing image data of the target person, and the service platform module is used to receive and display the printing model data and adjust the printing position to generate the preferred image printing data. The processing process of the data processing module also includes: Acquire the printing model data, adjust the printing model data to the position of the first type of image data, make two mutually perpendicular virtual planes through the midline of the face, namely virtual plane M1 and virtual plane M2, and virtual plane M1 is the median vertical plane of the printing model data; Randomly acquire a first monitoring point G1 and a second monitoring point G2 on the virtual plane M2, wherein the first monitoring point G1 and the second monitoring point G2 are symmetrically arranged with respect to the virtual plane M1; Draw a first perpendicular line through the first monitoring point G1 to the virtual plane M2, and the intersection of the first perpendicular line and the printed model is point G3; A second perpendicular line is drawn through the second monitoring point G2 to the virtual plane M2, and the intersection of the second perpendicular line and the printed model is point G4; Collect the dimension data Lg1 from the first monitoring point G1 to the point G3, and collect the dimension data Lg2 from the second monitoring point G2 to the point G4; Calculate the difference K between the size data Lg1 and the size data Lg2. The specific calculation process is: ； When the preset threshold Q6 < the difference degree K < the preset threshold Q7, the average value of the size data Lg1 and the size data Lg2 is obtained. ; Adjust the size data Lg1 and size data Lg2 to To generate preferred printing model data and send the preferred printing model data to the service platform module.

8. The image processing and printing system according to claim 7, characterized in that: The generation process of the preferred printing model data also includes: Import printing model data; Obtaining the location area of ​​facial features in the printing model data; When any one of the points G3 or G4 is located in the facial feature position area in the print model data, the size data Lg1 and the size data Lg2 are retained.

Images