CN112883833B - Body measuring and cutting method based on photographing and automatic body type identification - Google Patents
Body measuring and cutting method based on photographing and automatic body type identification Download PDFInfo
- Publication number
- CN112883833B CN112883833B CN202110127078.XA CN202110127078A CN112883833B CN 112883833 B CN112883833 B CN 112883833B CN 202110127078 A CN202110127078 A CN 202110127078A CN 112883833 B CN112883833 B CN 112883833B
- Authority
- CN
- China
- Prior art keywords
- line
- human body
- chest
- waist
- profile
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 238000005520 cutting process Methods 0.000 title claims description 12
- 238000005286 illumination Methods 0.000 claims abstract description 45
- 210000001217 buttock Anatomy 0.000 claims abstract description 43
- 238000012545 processing Methods 0.000 claims abstract description 10
- 210000001015 abdomen Anatomy 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 6
- 210000000689 upper leg Anatomy 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 230000000694 effects Effects 0.000 claims description 3
- 230000005484 gravity Effects 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 3
- 238000005259 measurement Methods 0.000 abstract description 12
- 230000037237 body shape Effects 0.000 description 9
- 210000003195 fascia Anatomy 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 238000007493 shaping process Methods 0.000 description 3
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
-
- A—HUMAN NECESSITIES
- A41—WEARING APPAREL
- A41H—APPLIANCES OR METHODS FOR MAKING CLOTHES, e.g. FOR DRESS-MAKING OR FOR TAILORING, NOT OTHERWISE PROVIDED FOR
- A41H1/00—Measuring aids or methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
- G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
Abstract
The invention discloses a body size measuring and tailoring method based on photographing and automatic body size identification, which comprises the following steps: s1) obtaining the height of a user, the whole body illumination of the back of the user and the whole body illumination of the side of the user; s2) performing image processing on the back whole body illumination and the side whole body illumination to acquire a human body back profile and a human body side profile of a user, and enabling the heights of the human body back profile and the human body side profile to be consistent; s3) finding out chest protruding points, waist back concave points and buttock protruding points in the side profile of the human body to serve as human body characteristic points, and calculating chest circumference, waistline and buttock circumference data of the human body according to the human body profile and the human body characteristic points. The body size measuring and tailoring method based on photographing and automatic body size identification can accurately acquire the characteristic points of the human body, thereby ensuring the measurement accuracy of key data such as chest circumference, waistline, hip circumference and the like, meeting the personalized body size measuring and tailoring requirements and greatly improving the use experience of users.
Description
Technical Field
The invention relates to an automatic body-shaping and tailoring method, in particular to a body-shaping and tailoring method based on photographing and automatic body shape identification.
Background
The traditional human body type data are mainly obtained by adopting a manual measurement mode, and specific data of human body parts such as three circumferences, shoulder widths, sleeve lengths and the like are obtained through a clothes ruler. With the continuous development of the internet and electronic commerce, automatic body-cutting measurements have also begun to be applied. For example, the depth sensing configuration of the human body 3D scanner can run around the body by six different angles and seven different heights, the size is fully and automatically measured to be more than 420, and the body contour of the user can be scanned by scanning the user transversely and longitudinally; however, the equipment has high manufacturing cost and large volume, and is only suitable for a small number of clothing customization enterprises which require customers to measure at appointed places and can bear high cost; moreover, the user can obtain data by holding or standing in the range of the scanner, remote measurement cannot be realized, and popularization and use of the scanner are further limited.
Some clothing customization merchants take photos by themselves, and acquire three customers through an automatic image identification processing mode, for example, a method for realizing the body size measurement of clothing customization through a photo in patent document CN201610026297.8 cannot guarantee whether the photo is taken correctly, so calculation errors are caused, the problem that how to judge the position of body type data of the human body is solved, human body measurement still needs to be realized by means of manual image reading, meanwhile, the patent does not give out a specific calculation process and result of an experience value, a plurality of adverse factors threaten the calculation of the experience value, and the error value between an estimated value and an actual measurement is difficult to ensure to be within a reasonable range, so improvement is urgently needed.
The method for remotely customizing garments disclosed in patent document CN201180023379.2, sends two his/her profile photos to a processing device that employs zooming the photos to the human body 1:1 to determine the required clothing size or clothing size.
Further, for example, in a system method for automatically calculating a human body shape by photographing disclosed in patent document CN201710058112.6, characteristic point data of the human body shape is obtained according to pixel points occupied by a preset line segment, so as to determine a required garment size or a required garment size.
From the above, the existing automatic body-shaping method adopts simple contour extraction and proportion conversion to estimate three-dimensional data of human body, so that not only is the estimation error larger, but also the human body engineering curve is not met, and especially the individualized data identification of key human body characteristic point data such as chest circumference, waistline, hip circumference and the like is not accurate enough, and the body type difference of different people cannot be considered. In addition, because the human body curve is complex and the individuals are very different, the error of the image recognition processing is also large, the prior art generally needs to judge the position of the human body shape data through the manual image reading of a professional with body measurement experience, and then manually calculate the body shape data or correct the automatic measurement data, so that the personalized body shape cutting requirement can be finally met, and the whole process also often generates errors due to photographing angle problems, light problems and the like. The superposition of various errors finally seriously affects the measurement precision, and greatly affects the use experience of users.
Disclosure of Invention
The invention aims to solve the technical problem of providing a body size cutting method based on photographing and automatic body type identification, which can accurately acquire human body characteristic points, thereby ensuring the measurement accuracy of key data such as chest circumference, waistline, hip circumference and the like, meeting the personalized body size cutting requirements and greatly improving the use experience of users.
The invention provides a body measuring and tailoring method based on photographing and automatic body type identification, which aims to solve the technical problems and comprises the following steps: s1) obtaining the height of a user, the whole body illumination of the back of the user and the whole body illumination of the side of the user; s2) performing image processing on the back whole body illumination and the side whole body illumination to acquire a human body back profile and a human body side profile of a user, and enabling the heights of the human body back profile and the human body side profile to be consistent; s3) finding out chest protruding points, waist back concave points and buttock protruding points in the side profile of the human body to serve as human body characteristic points, and calculating chest circumference, waistline and buttock circumference data of the human body according to the human body profile and the human body characteristic points.
The above-mentioned body type measuring and tailoring method based on photographing automatic body type identification, wherein the step S1 includes: s11: placing the mobile phone on the wall, wherein the included angle between the mobile phone and the ground is alpha, and the included angle alpha is calculated through self gravity induction of the mobile phone; s12: the user keeps a certain distance from the mobile phone, so that the whole body of the user can be shot by the mobile phone lens; s13: the user is opposite to the mobile phone, two feet are gathered, arms are opened to form a certain angle with the body, the angle is 15-45 degrees, and under voice prompt, the whole body illumination collection on the back of the user is completed; s14: the side face of the user faces the mobile phone, the two hands and the feet are gathered up and stand straight, and under the voice prompt, the collection of the whole body illumination of the side face of the user is completed.
The above-mentioned body type measuring and tailoring method based on photographing automatic body type identification, wherein the step S2 includes: and (2) acquiring an included angle alpha in the step S1, and stretching and changing the back whole body illumination and the side whole body illumination according to the included angle alpha, so as to eliminate human deformation errors caused by shooting angles and enable the heights of users in the back whole body illumination and the side whole body illumination to be consistent.
The above-mentioned body type measuring and tailoring method based on photographing automatic body type identification, wherein the step S3 includes: s31: finding out the most protruding point of the chest in the profile of the side surface of the human body, positioning the side chest line, finding out the equal height position in the profile of the back surface of the human body according to the height ratio between the top of the head and the heel of the side chest line, namely, the back chest line, and calculating the actual lengths of the side chest line and the back chest line according to the height of the user; estimating chest circumference according to the actual lengths of the lateral chest lines and the back chest lines; s32: finding out the most concave point of the waist and the back in the profile of the side surface of the human body, positioning the side waist line, finding out the equal-height position in the profile of the back surface of the human body according to the height ratio between the top of the head and the heel of the side waist line, namely, obtaining the back waist line, calculating the actual lengths of the side waist line and the back waist line according to the height of a user, and estimating the waistline according to the actual lengths of the side waist line and the back waist line; s33: in the profile of the human body side, the most prominent point of buttocks is found, the side buttocks line is positioned, the equal height position is found in the profile of the human body back according to the height proportion between the head top and the heel of the side buttocks line, namely the back buttocks line, the actual lengths of the side buttocks line and the back buttocks line are calculated according to the height, and the hip circumference is estimated according to the actual lengths of the side buttocks line and the back buttocks line.
The above-mentioned body type measuring and tailoring method based on photographing automatic body type identification, wherein the step S3 includes: s34: acquiring the head top positions of the back profile and the side profile of the human body, and determining a head top line; s35: acquiring the positions of foot roots of the back profile and the side profile of the human body, and determining foot root lines, wherein the foot root lines are parallel to the top line; s36: the chest girth is led out through the most protruding point of the chest, the chest girth is parallel to the head top line, the line segment of the chest girth passing through the contour of the back of the human body is the back chest line, and the line segment of the chest girth passing through the contour of the side of the human body is the side chest line; s37: the waist line is led out through the most protruding point of the waist, the waist line is parallel to the head top line, the line segment of the waist line passing through the contour of the back of the human body is the back waist line, and the line segment of the waist line passing through the contour of the side surface of the human body is the side waist line; s38: the hip line is led out through the most protruding point of the hip, the hip line is parallel to the top line, the line segment of the hip line passing through the contour of the back of the human body is the back hip line, and the line segment of the hip line passing through the contour of the side of the human body is the side hip line.
The above-mentioned body type measuring and tailoring method based on photographing automatic body type identification, wherein, the step S3 further comprises: the proportionality coefficient is determined according to the distance between the overhead line L1 and the foot line L2 and the inputted height of the user, the chest circumference=the empirical value×the proportionality coefficient×1.57× (back chest line+side chest line), the waist circumference=the empirical value×the proportionality coefficient×1.57× (back waist line+side waist line), and the hip circumference=the empirical value×the proportionality coefficient×1.57× (back hip line+side hip line).
The above body size measuring and tailoring method based on photographing and automatic body size identification, wherein the step S1 further comprises the step of obtaining the whole body illumination of the front face of the user, the step S2 performs image processing on the whole body illumination of the front face of the user, the front face profile of the human body is obtained, the height of the front face profile of the human body is consistent with the height of the side profile of the human body, the line segment of the chest circumference line passing through the front face profile of the human body in the step S3 is a positive chest line, the line segment of the waistline line passing through the front face profile of the human body is a positive waist line, and the line segment of the hip circumference line passing through the front face profile of the human body is a positive hip line; the positive chest line is used for correcting the back chest line, the positive waist line is used for correcting the back waist line, and the positive hip line is used for correcting the back hip line.
According to the body size measuring and tailoring method based on photographing and automatic body type identification, the correction process of the back chest line is as follows: if the length difference delta L between the positive chest line and the back chest line exceeds a first preset threshold value, the most prominent point of the chest originally found in the profile of the human body is moved upwards by n x delta L, n is a natural number, n takes a value of 1-3, the positive chest line and the back chest line are redetermined, and if the length of the new back chest line is greater than that of the original back chest line and the length difference delta L between the new positive chest line and the new back chest line is within the first preset threshold value, the new back chest line is directly adopted as the corrected back chest line.
The method for measuring and tailoring the body shape based on photographing and automatic identification further comprises the steps of establishing a clothing layout library in advance, calculating chest circumference, waistline and hip circumference data of a human body according to the height of a user, selecting matched clothing from the clothing layout library to be overlapped outside the outline of the human body, and giving out a try-on effect display diagram; the step S2 includes identifying contour points of the garment in the back whole body photograph and the side whole body photograph, determining the type of the garment according to the contour points of the garment, peeling the human body from the photo background by using an example segmentation technology, and removing the garment to obtain a human body back contour and a human body side contour corresponding to the net size of the human body.
The body type auxiliary characteristic points comprise neck points, armpit points, abdomen high points, shoulder neck points, waist high points and thigh root points, the neck circumference, shoulder width, abdomen circumference, thigh root circumference, front waist section, rear waist section, front chest width, rear back width, arm circumference, armholes, cuffs and sleeve lengths are determined according to the body type auxiliary characteristic points in the step S3, and correction data are given by comparing with clothes selected from a clothes layout library.
Compared with the prior art, the invention has the following beneficial effects: according to the body size measuring and tailoring method based on photographing and automatic body size identification, the human body characteristic points are quickly found by matching the human body back profile and the human body side profile which are consistent in height, and the chest circumference, waistline and hip circumference data of the human body are calculated, so that key human body characteristic points can be accurately identified and acquired, accurate measurement and correction can be conveniently and quickly carried out on the chest circumference, waistline, hip circumference and other key data on the same plane, final estimation accuracy is improved, use experience of a user is greatly improved, and personalized body size and tailoring requirements are met.
Drawings
FIG. 1 is a schematic diagram of a body type cutting process based on photographing automatic body type identification in the invention;
FIG. 2 is a schematic view of a whole body of a back side taken by a mobile phone according to the present invention;
FIG. 3 is a schematic diagram of a side whole body shot by a mobile phone according to the invention;
FIG. 4a is a side original photograph, FIG. 4b is a side restored photograph, used in the present invention;
FIG. 5a is a back side original photograph, FIG. 5b is a back side restored photograph, used in the present invention;
FIG. 6 is a schematic illustration of the determination of the back chest line and side chest line using body contours in accordance with the present invention;
FIG. 7 is a schematic view of the present invention for determining the back and side fascia using the contours of the human body;
FIG. 8 is a schematic view of the present invention for determining the back and side hip lines using the contours of the human body.
Detailed Description
The invention is further described below with reference to the drawings and examples.
Fig. 1 is a schematic diagram of a body shape measuring and cutting process based on photographing automatic body shape identification.
Referring to fig. 1, the body size cutting method based on photographing automatic body type identification provided by the invention comprises the following steps:
s1) obtaining the height of a user, the whole body illumination of the back of the user and the whole body illumination of the side of the user;
s2) performing image processing on the back whole body illumination and the side whole body illumination to acquire a human body back profile and a human body side profile of a user, and enabling the heights of the human body back profile and the human body side profile to be consistent;
s3) finding out the most protruding points of the chest, the most concave points of the waist and the back and the most protruding points of the buttocks in the profile of the side of the human body to be used as the characteristic points of the human body, and calculating the chest circumference, waistline and hip circumference data of the human body according to the profile of the human body and the characteristic points of the human body.
When the invention is used, the mobile phone is obliquely arranged at the basement, meanwhile, the front camera of the mobile phone is called to shoot the whole body of the back side and the whole body of the side of the human body, the height data of the body is uploaded to the system, the human body is peeled off from the photo background by utilizing an example segmentation technology, the position of the body type data is positioned according to the contour of the human body and the requirements of the clothing industry, and finally, the data of the chest circumference, waistline, hip circumference and the like of the human body are calculated, and the invention comprises the following steps:
step 1, inputting the height of a user;
step 2, the mobile phone is obliquely arranged at a corner, and the angle alpha between the mobile phone and the ground is calculated by utilizing the gravity induction of the mobile phone;
step 3, the user needs to wear tight clothes and keep a certain distance from the mobile phone until the whole body can be shot by the lens of the mobile phone;
step 4, the user faces away from the mobile phone, the mobile phone voice prompts the user to open the arm to form 30 degrees with the upper body, and the user gathers the feet;
step 5, the mobile phone broadcasts the countdown by voice, so that a user is guided to keep a photographing posture, and back side photographing is completed, as shown in fig. 2;
step 6, the mobile phone voice prompts the user to turn right, the side faces face the mobile phone, and the two hands and the feet are gathered up and stand straight;
step 7, the mobile phone performs voice broadcasting countdown, guides the user to keep a photographing posture, and completes side photographing, as shown in fig. 3;
step 8, adjusting an original photo according to the angle alpha between the mobile phone and the ground to eliminate errors caused by oblique angles, wherein FIG. 4a is an original side photo used in the invention, and FIG. 4b is a restored side photo used in the invention; FIG. 5a is a back side original photograph, FIG. 5b is a back side restored photograph, used in the present invention; therefore, the mobile phone is not required to be perpendicular to the ground, the convenience of use of a user can be improved, and the recognition accuracy can be ensured.
Step 9, after obtaining back and side photos, peeling the human body from the complex background by using an example segmentation technology to obtain the back and side contours of the human body;
step 10, finding out the most protruding point of the chest in the profile of the human body, positioning the lateral chest line, finding out the equal height position in the profile of the human body on the back according to the height ratio between the top of the head and the heel of the lateral chest line, namely, the back chest line, calculating the actual lengths of the lateral chest line and the back chest line according to the height, setting the lateral chest line and the back chest line as the length and the width of a rectangle, calculating the perimeter of the rectangle inscribed ellipse, and finally obtaining chest circumference = experience value multiplied by 1.57× (back chest line+lateral chest line) according to an ellipse formula; the typical male chest circumference experience value is 1.029, and the female chest circumference experience value is 1.072;
step 11, finding out the most concave point of the waist and the back in the profile of the human body, positioning the side waist line, finding out the equal height position in the profile of the human body on the back according to the height ratio between the top of the head and the heel of the side waist line, namely, the back waist line, calculating the actual lengths of the side waist line and the back waist line according to the height, setting the side waist line and the back waist line as the length and the width of a rectangle, calculating the perimeter of the rectangle inscribed ellipse, and finally obtaining the waistline=experience value multiplied by 1.57× (back waist line+side waist line) according to an ellipse formula; the waistline experience value of a general man and a female is 1.012;
step 12, finding out the most prominent point of buttocks in the profile of the human body, positioning a side buttock line, finding out the equal height position in the profile of the human body on the back according to the height ratio between the head top and the heel of the side buttock line, namely a back buttock line, calculating the actual lengths of the side buttock line and the back buttock line according to the height, setting the side buttock line and the back buttock line as the length and the width of a rectangle, calculating the perimeter of the rectangle inscribed ellipse, and finally obtaining the buttock = empirical value multiplied by 1.57× (back buttock line+side buttock line) according to an ellipse formula; typically, the hip circumference of a male or female has an empirical value of 1.053.
With continued reference to fig. 6-8, to ensure that the contour position is found quickly and accurately, the following steps are performed:
1) Acquiring the head top positions of the back profile and the side profile of the human body, and determining a head top line L1;
2) Acquiring foot positions of a back profile and a side profile of a human body, determining a foot line L2, wherein the foot line L2 is parallel to a head top line L1, and determining a proportionality coefficient according to the distance between the head top line L1 and the foot line L2 and the input height of a user;
3) The chest girth line L3 is led out through the most protruding point A of the chest, the chest girth line L3 is parallel to the head and top line L1 and the foot root line L2, and the chest girth line L3 passes through the line segment L of the back contour of the human body bx Namely a back chest line, the chest line L3 passes through a line segment L of the side contour of the human body cx Namely a lateral chest line, as shown in fig. 6; chest circumference = empirical x 1.57 x scaling factor x (L bx +L cx );
4) A waistline L4 is led out through a point B with the most protruding waist, the waistline L4 is parallel to a vertex line L1, and the waistline L4 passes through a line segment L of the contour of the back of a human body by Namely a back waist line, the waist line L4 passes through a line segment L of the side contour of the human body cy Namely, a side waist line, as shown in fig. 7; waistline = empirical x 1.57 x scaling factor x (L by +L cy );
5) A hip-surrounding line L5 is led out from the most protruding point C of the buttocks, the hip-surrounding line L5 is parallel to the head-top line L1, and the hip-surrounding line L5 passes through a line segment L of the back contour of the human body bt I.e. the back-hip line, the hip line L5 passes through the line segment L of the human body side contour ct Namely a side hip line, as shown in fig. 8; hip circumference = empirical x 1.57 x scaling factor×(L bt +L ct )。
The invention uses the contour of the back of the human body to match with the contour of the side surface of the human body to calculate the data of the chest circumference, the waistline and the hip circumference of the human body, and does not directly use the contour of the front of the human body, and the main reason is that:
1) If the front whole body illumination is directly used, the privacy of the user is difficult to protect, and the back whole body illumination can be shot to more accurately acquire the height and key characteristic points of the human body, and the front whole body illumination can easily and incorrectly identify the instep part as the height part; the invention adopts the back whole body illumination to replace the front whole body illumination, firstly can protect the privacy of a user, secondly can clear the influence of instep parts on height identification, thirdly can find out the lateral chest, waist and hip lines horizontally aligned to the back whole body illumination through the heel and head top positions of the back and the lateral heel and head top pair Ji Beimian whole body illumination and the lateral whole body illumination, and find out the corresponding back chest, waist and hip lines on the same horizontal plane by finding out the lateral chest, waist and hip lines on the lateral whole body illumination, and ensuring that the back chest lines are on the same horizontal plane as the lateral chest lines, the back waist lines are on the same horizontal plane as the lateral waist lines.
2) If the longest line is found in the chest area of the front and the side respectively as chest line, it is found in practice that the side chest line and the front chest line found by this method are often not in the same horizontal plane, and the longest line found by the front chest is basically in the armpit position and not on the chest circumference.
3) If the abdomen is judged to be convex (or concave), the longest (or shortest) line is found in the waist regions of the front and the side surfaces respectively to be used as the waist line, in practical application, a plurality of lines with equal length are found in the abdomen of the side surface for people with flat abdomen, which is the side waist line is difficult to determine, and the side waist line and the front waist line found by the method are often not in the same horizontal plane.
4) If the longest lines are found in the buttocks of the front and the side respectively, namely the buttocks lines, the side buttocks lines and the front buttocks lines found by the method are often not in the same horizontal plane, and the longest lines found by the front chest are basically in the armpit position and are not on the chest circumference.
The invention does not directly use the front outline of the human body after the front whole body illumination treatment to determine the human body key data, but can consider the use of the front outline of the human body to correct the measured human body key data. Specifically, the step S1 of the present invention further includes obtaining a front whole body illumination of the user, the step S2 performs image processing on the front whole body illumination to obtain a front contour of the human body of the user, and the heights of the front contour of the human body and the side contour of the human body are identical, a line segment of the chest circumference line passing through the front contour of the human body in the step S3 is a positive chest line, a line segment of the waistline line passing through the front contour of the human body is a positive waist line, and a line segment of the hip circumference line passing through the front contour of the human body is a positive hip line; the positive chest line is used for correcting the back chest line, and if the difference value delta L between the positive chest line and the back chest line exceeds a preset threshold value, the user is reminded to input back whole body illumination again; if the difference between the positive chest line and the negative chest line is within the preset threshold, the average value of the positive chest line and the negative chest line can be taken or not processed. Likewise, the positive fascia is used to correct the back fascia and the positive fascia is used to correct the back fascia.
The determination of the protruding points of the chest is relatively complex, and the found most protruding points of the chest in practical application tend to be close to the armpit position, and the invention further provides the following automatic correction process of the back chest line: if the difference delta L between the lengths of the positive chest line and the back chest line exceeds a first preset threshold, the most prominent point of the chest originally found in the profile of the human body is moved upwards by n x delta L, n is a natural number, preferably 1-3, the positive chest line and the back chest line are redetermined, and if the length of the new back chest line is greater than the length of the original back chest line and the difference delta L between the lengths of the new positive chest line and the new back chest line is within the first preset threshold, the new back chest line is directly used as the corrected back chest line. In addition, the invention can be further fine-tuned to ensure that the found chest protrusion point is the most protruding point of the chest, and the specific fine-tuning process is as follows: if the difference between the lengths of the positive chest line and the negative chest line is within a first preset threshold value but is larger than a second preset threshold value, the second preset threshold value is smaller than the first preset threshold value, the positive chest line and the negative chest line are moved upwards at a certain interval, the positive chest line and the negative chest line are determined again, and if the length of the new negative chest line is larger than the length of the original negative chest line, the positive chest line and the negative chest line are moved upwards at the same interval until the new negative chest line is not lengthened any more; and taking the longest back chest line as the corrected back chest line if the length difference delta L between the longest back chest line and the corresponding positive chest line is within a second preset threshold value.
The recognition error of the lumbar and dorsal pits or the buttocks prominences is relatively small, so that the correction process of the waist line and the buttocks line preferably selects the average value of the two or does not process. If the adjustment is needed, so as to find the most concave point of the waist and back or the most prominent point of the buttocks, the user needs to move in the up-down direction to find the longest back waist line or back and buttocks line. Alternatively, for the belt line, if the length of the positive belt line is greater than the back belt line, it is preferable to move upward to find the longest back belt line, whereas if the length of the positive belt line is less than the back belt line, it is preferable to move downward to find the longest back belt line. The same is true for the hip line, and will not be described in detail herein.
Further, the invention also comprises the steps of pre-establishing a clothing layout library, calculating the chest circumference, waistline and hip circumference data of the human body according to the height of the user, selecting matched clothing from the clothing layout library to be overlapped outside the outline of the human body, and giving out a try-on effect display diagram. The step S2 of the invention comprises the steps of identifying contour points of the clothing in the back whole body photograph and the side whole body photograph, determining the type of the clothing according to the contour points of the clothing, peeling the human body from the photo background by using an example segmentation technology, and removing the clothing to obtain the contour of the back surface of the human body and the contour of the side surface of the human body corresponding to the net size of the human body.
Further, the step S3 of the present invention further includes selecting body type auxiliary feature points on the human body contour, the body type auxiliary feature points including neck points, armpit points, abdomen high points, shoulder neck points, waist high points and thigh root points, and the step S3 determines neck circumference, shoulder width, abdomen circumference, thigh root circumference, front waist section, rear waist section, front chest width, rear back width, arm circumference, armholes, cuffs and sleeve lengths according to the body type auxiliary feature points, and compares them with the clothing selected from the clothing layout library to give correction data.
While the invention has been described with reference to the preferred embodiments, it is not intended to limit the invention thereto, and it is to be understood that other modifications and improvements may be made by those skilled in the art without departing from the spirit and scope of the invention, which is therefore defined by the appended claims.
Claims (5)
1. The body size measuring and tailoring method based on photographing and automatic body size identification is characterized by comprising the following steps of:
s1) obtaining the height of a user, the whole body illumination of the back of the user and the whole body illumination of the side of the user;
s2) performing image processing on the back whole body illumination and the side whole body illumination to acquire a human body back profile and a human body side profile of a user, and enabling the heights of the human body back profile and the human body side profile to be consistent;
s3) finding out chest protruding points, waist back concave points and buttock protruding points in the side profile of the human body to serve as human body characteristic points, and calculating chest circumference, waistline and buttock circumference data of the human body according to the human body profile and the human body characteristic points;
the step S1 includes:
s11: placing the mobile phone on the wall, wherein the included angle between the mobile phone and the ground is alpha, and the included angle alpha is calculated through self gravity induction of the mobile phone;
s12: the user keeps a certain distance from the mobile phone, so that the whole body of the user can be shot by the mobile phone lens;
s13: the user is opposite to the mobile phone, two feet are gathered, arms are opened to form a certain angle with the body, the angle is 15-45 degrees, and under voice prompt, the whole body illumination collection on the back of the user is completed;
s14: the side face of the user faces the mobile phone, the two hands and the feet are gathered up and stand straight, and under the voice prompt, the collection of the whole body illumination of the side face of the user is completed;
the step S2 includes: acquiring an included angle alpha in the step S1, and stretching and changing the back whole body illumination and the side whole body illumination according to the included angle alpha to eliminate human deformation errors caused by shooting angles and enable the heights of users in the back whole body illumination and the side whole body illumination to be consistent;
the step S3 includes:
s31: finding out the most protruding point of the chest in the profile of the side surface of the human body, positioning the side chest line, finding out the equal height position in the profile of the back surface of the human body according to the height ratio between the top of the head and the heel of the side chest line, namely, the back chest line, and calculating the actual lengths of the side chest line and the back chest line according to the height of the user; estimating chest circumference according to the actual lengths of the lateral chest lines and the back chest lines;
s32: finding out the most concave point of the waist and the back in the profile of the side surface of the human body, positioning the side waist line, finding out the equal-height position in the profile of the back surface of the human body according to the height ratio between the top of the head and the heel of the side waist line, namely, obtaining the back waist line, calculating the actual lengths of the side waist line and the back waist line according to the height of a user, and estimating the waistline according to the actual lengths of the side waist line and the back waist line;
s33: finding out the most prominent point of buttocks in the profile of the side face of the human body, positioning a side buttock line, finding out the equal height position in the profile of the back face of the human body according to the height ratio between the head top and the heel of the side buttock line, namely a back buttock line, calculating the actual lengths of the side buttock line and the back buttock line according to the height, and estimating the hip circumference according to the actual lengths of the side buttock line and the back buttock line;
the step S1 further comprises the step of obtaining a front whole body illumination of the user, the step S2 is used for carrying out image processing on the front whole body illumination to obtain a front outline of the human body of the user, the heights of the front outline of the human body and the side outline of the human body are consistent, a line segment of a chest girth passing through the front outline of the human body in the step S3 is a positive chest line, a line segment of a waistline passing through the front outline of the human body is a positive waist line, and a line segment of a hip girth passing through the front outline of the human body is a positive hip line; the positive chest line is used for correcting the back chest line, the positive waist line is used for correcting the back waist line, and the positive hip line is used for correcting the back hip line;
the correction process of the back chest line is as follows:
if the length difference delta L between the positive chest line and the back chest line exceeds a first preset threshold value, the most prominent point of the chest originally found in the profile of the human body is moved upwards by n x delta L, n is a natural number, n takes a value of 1-3, the positive chest line and the back chest line are redetermined, and if the length of the new back chest line is greater than that of the original back chest line and the length difference delta L between the new positive chest line and the new back chest line is within the first preset threshold value, the new back chest line is directly adopted as the corrected back chest line.
2. The method for body size cutting based on photographing automatic body type recognition according to claim 1, wherein the step S3 comprises:
s34: acquiring the head top positions of the back profile and the side profile of the human body, and determining a head top line;
s35: acquiring the positions of foot roots of the back profile and the side profile of the human body, and determining foot root lines, wherein the foot root lines are parallel to the top line;
s36: the chest girth is led out through the most protruding point of the chest, the chest girth is parallel to the head top line, the line segment of the chest girth passing through the contour of the back of the human body is the back chest line, and the line segment of the chest girth passing through the contour of the side of the human body is the side chest line;
s37: the waist line is led out through the most protruding point of the waist, the waist line is parallel to the head top line, the line segment of the waist line passing through the contour of the back of the human body is the back waist line, and the line segment of the waist line passing through the contour of the side surface of the human body is the side waist line;
s38: the hip line is led out through the most protruding point of the hip, the hip line is parallel to the top line, the line segment of the hip line passing through the contour of the back of the human body is the back hip line, and the line segment of the hip line passing through the contour of the side of the human body is the side hip line.
3. The method for body size cutting based on photographing automatic body type recognition according to claim 2, wherein the step S3 further comprises: the proportionality coefficient is determined according to the distance between the overhead line L1 and the foot line L2 and the inputted height of the user, the chest circumference=the empirical value×the proportionality coefficient×1.57× (back chest line+side chest line), the waist circumference=the empirical value×the proportionality coefficient×1.57× (back waist line+side waist line), and the hip circumference=the empirical value×the proportionality coefficient×1.57× (back hip line+side hip line).
4. The method for measuring and tailoring clothes based on photographing and automatically recognizing body types according to claim 1, further comprising the steps of establishing a clothes layout library in advance, calculating chest circumference, waistline and hip circumference data of a human body according to the height of a user, selecting matched clothes from the clothes layout library to be overlapped outside the contour of the human body, and giving out a try-on effect display diagram; the step S2 includes identifying contour points of the garment in the back whole body photograph and the side whole body photograph, determining the type of the garment according to the contour points of the garment, peeling the human body from the photo background by using an example segmentation technology, and removing the garment to obtain a human body back contour and a human body side contour corresponding to the net size of the human body.
5. The method for body size cutting based on photographing automatic body size recognition according to claim 4, wherein the step S3 further comprises selecting body size auxiliary feature points on the contour of the human body, the body size auxiliary feature points including neck points, armpit points, abdomen high points, shoulder neck points, waist high points and thigh root points, and the step S3 determines neck circumference, shoulder width, abdomen circumference, thigh root circumference, front waist section, rear waist section, front chest width, rear back width, arm circumference, armholes, cuffs and sleeve length according to the body size auxiliary feature points, and compares the body size auxiliary feature points with the selected clothes from the clothes layout library to give correction data.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110127078.XA CN112883833B (en) | 2021-01-29 | 2021-01-29 | Body measuring and cutting method based on photographing and automatic body type identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110127078.XA CN112883833B (en) | 2021-01-29 | 2021-01-29 | Body measuring and cutting method based on photographing and automatic body type identification |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112883833A CN112883833A (en) | 2021-06-01 |
CN112883833B true CN112883833B (en) | 2023-05-05 |
Family
ID=76053702
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110127078.XA Active CN112883833B (en) | 2021-01-29 | 2021-01-29 | Body measuring and cutting method based on photographing and automatic body type identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112883833B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113628233A (en) * | 2021-08-02 | 2021-11-09 | 嘉兴九全科技服务有限公司 | Internet-based body type management method, device, equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106666902A (en) * | 2017-01-23 | 2017-05-17 | 陈婕 | Systematic method for automatically calculating human body shape by taking pictures |
CN107240007A (en) * | 2017-07-21 | 2017-10-10 | 陕西科技大学 | A kind of AR three-dimensional virtual fitting systems combined with 3D manikins |
CN110151186A (en) * | 2019-05-28 | 2019-08-23 | 北京智形天下科技有限责任公司 | A kind of human body measurement method based on network-enabled intelligent terminal |
CN110188726A (en) * | 2019-06-05 | 2019-08-30 | 杭州智珺智能科技有限公司 | Online human body dimension measurement method and the amount body clothing for this method |
WO2020141565A1 (en) * | 2018-12-31 | 2020-07-09 | I-Deal S.R.L. | Anthropometric data portable acquisition device and method of collecting anthropometric data |
CN112116702A (en) * | 2020-09-01 | 2020-12-22 | 上海云杉服装科技有限公司 | Human body measurement three-dimensional modeling method and system based on front and side profiles |
-
2021
- 2021-01-29 CN CN202110127078.XA patent/CN112883833B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106666902A (en) * | 2017-01-23 | 2017-05-17 | 陈婕 | Systematic method for automatically calculating human body shape by taking pictures |
CN107240007A (en) * | 2017-07-21 | 2017-10-10 | 陕西科技大学 | A kind of AR three-dimensional virtual fitting systems combined with 3D manikins |
WO2020141565A1 (en) * | 2018-12-31 | 2020-07-09 | I-Deal S.R.L. | Anthropometric data portable acquisition device and method of collecting anthropometric data |
CN110151186A (en) * | 2019-05-28 | 2019-08-23 | 北京智形天下科技有限责任公司 | A kind of human body measurement method based on network-enabled intelligent terminal |
CN110188726A (en) * | 2019-06-05 | 2019-08-30 | 杭州智珺智能科技有限公司 | Online human body dimension measurement method and the amount body clothing for this method |
CN112116702A (en) * | 2020-09-01 | 2020-12-22 | 上海云杉服装科技有限公司 | Human body measurement three-dimensional modeling method and system based on front and side profiles |
Also Published As
Publication number | Publication date |
---|---|
CN112883833A (en) | 2021-06-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107292948B (en) | Human body modeling method and device and electronic equipment | |
KR102132721B1 (en) | Method, server and program of acquiring image for measuring a body size and a method for measuring a body size using the same | |
CN107273846B (en) | Human body shape parameter determination method and device | |
US8813378B2 (en) | System and method for drafting garment patterns from photographs and style drawings | |
JP6387417B2 (en) | Portable information terminal | |
US8571698B2 (en) | Simple techniques for three-dimensional modeling | |
CN103535960B (en) | Human body three-dimensional measurement method based on digital images | |
CN106659259A (en) | Method for virtually selecting clothing | |
US9715759B2 (en) | Reference object for three-dimensional modeling | |
CN106666902A (en) | Systematic method for automatically calculating human body shape by taking pictures | |
CN112883833B (en) | Body measuring and cutting method based on photographing and automatic body type identification | |
US20180247426A1 (en) | System for accurate remote measurement | |
CN105956910A (en) | Clothes customized method, image shooting device and control system | |
CN112116702A (en) | Human body measurement three-dimensional modeling method and system based on front and side profiles | |
US20180096490A1 (en) | Method for determining anthropometric measurements of person | |
KR20150011907A (en) | Apparatus and Method for generating user's three dimensional body model based on depth information | |
JPH0944556A (en) | Trial wearing simulation method | |
CN114375177A (en) | Body measurement device and control method thereof | |
JP2013228334A (en) | Three-dimensional measuring system, three-dimensional measuring method and three-dimensional measuring program | |
KR20180007016A (en) | calculating bodysize system using two photos | |
Sehgal et al. | Automatic Extraction of 3d body measurements from 2d images of a female form | |
CN109801326A (en) | It is a kind of for obtaining the image measuring method of human somatotype data | |
CN106942813A (en) | The contactless convenient measuring method for obtaining somatic data | |
CN108564586A (en) | A kind of body curve's measurement method and system based on deep learning | |
CN106296797A (en) | A kind of spatial digitizer characteristic point modeling data processing method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20231211 Address after: Room 305, Building 1, West 1160, No. 1687 Changyang Road, Yangpu District, Shanghai, 200090 Patentee after: Shanghai Delan Technology Co.,Ltd. Address before: 200433 3rd floor, No.1 Lane 127, Guotai Road, Yangpu District, Shanghai Patentee before: Shanghai Weima Intelligent Technology Co.,Ltd. |
|
TR01 | Transfer of patent right |