TWI421466B - Method for measurement of clothes sizes by video device - Google Patents

Description

Method for measuring the type of clothes worn by a video device

The invention relates to a method for measuring a type of clothes worn by a video device, in particular, a video device is used to capture images of various types of clothes and a customer's body shape, and then the arithmetic unit compares the two images to achieve fast. The convenience of screening the right size clothes for customers.

According to the non-contact optical measurement technology to examine the three-dimensional shape of the object to be tested, it has been widely used in bioengineering and precision manufacturing. It mainly uses the method of fringe projection to determine the shape of the object by using the amount of distortion of the stripe projected on the object to be tested.

The existing measurement stripe projection contour technology can be further divided into an image comparison mode and an absolute mode, both of which need to have a periodic stripe pattern projected onto the surface to be tested, and then demodulate to correspond to the twisted stripe of each depth. However, this measurement fringe projection contour technique is difficult to accurately demodulate when measuring the three-dimensional shape of a large-sized object to correspond to the twisted stripe of each depth.

Therefore, there is a three-dimensional shape measurement system which is developed in response to the three-dimensional shape of a large-sized object, and the three-dimensional shape measurement system mainly includes a reference standard piece for establishing a phase value of the reference coordinate, Stripe projection device for projecting a periodic light Learning stripe on the reference standard sheet, a moving platform for axially moving the reference standard sheet, an image sensing element for receiving an image of the reference standard sheet, and a phase calculating unit for storing and calculating the Referring to the phase value of the standard image; in use, after calculating the phase value of the reference coordinate of the reference standard by the phase calculation unit, placing a test object on the mobile platform, and receiving the image by the image sensing component And stored in the phase calculation unit, and then the phase calculation unit calculates the three-dimensional shape phase value of the object to be tested.

However, the existing three-dimensional shape measurement system is specifically used in the industry to detect the three-dimensional shape of the surface of the object to be tested, although the function is strong, but the equipment technology required is more professional, so it is not suitable for general business dealings, such as : The purchase of clothes is used in general life and life.

Therefore, the present inventors have in view of the fact that the existing measurement stripe projection contour technology and the three-dimensional shape measurement system are technically high on the device, and the operation is relatively difficult to be missing, etc., by virtue of its many years of manufacturing and design in related fields. With the help of experience and knowledge, and through a lot of ingenuity, the method of measuring the type of clothes worn by the video device of the present invention has been developed.

The invention relates to a method for measuring the type of clothes worn by the video device, and the main purpose of the invention is to provide an image capturing and comparing method for quickly screening the appropriate size clothes for the customer.

In order to achieve the above object, the inventors have studied the following method for measuring the type of clothes worn by the video device, and the implementation steps include: A. System correction: first correcting the distance between the video device and the backplane and the target, And correcting the pixels of the photographing back sheet and the target object, which is a model wearing clothes. One of the children or the customer; B. Video capture: the image of the target device is captured by the video device; C. pixel analysis: and the arithmetic unit performs pixel analysis on the image of the captured object; Gray-scale and binarization processing: Continue to make the operation unit first gray-scale the image of the target object, and then perform binarization processing to facilitate accurate capture of the object coordinates; E. draw clothes lines Contour: In turn, the video device is driven by the computing unit to input the captured image into the database of the computing unit according to the contour of each type of clothing worn by the model; F. capturing the contour of the customer: driving the video device through the computing unit according to Customer's body contour for image capture; G. Database comparison: Compare the contour of the customer's body shape with the contour of the clothing filed by the database; H. Suggested output: the model of the clothes that will suit the customer The display device displays the output.

By means of the simple equipment and convenient operation, the general customer can purchase the clothes without actually trying on the clothes, and only through the invention system can judge the suitable size clothes according to the customer's body shape, so that not only the saving can be achieved. When the customer purchases clothes, the time for the store to collect and organize the clothes can be reduced.

(1)‧‧‧Video devices

(2) ‧‧‧ arithmetic unit

(3)‧‧‧Display devices

(4)‧‧‧Models

(5) ‧‧ ‧ fixed point

(6)‧‧‧ Backboard

(61)‧‧‧Sensor unit

(7) ‧ ‧ customers

First figure: the architecture diagram of the present invention

Second figure: flow chart of the present invention

The third figure: the implementation step of the input garment size of the present invention (1)

Figure 4: Implementation of the input garment size step of the present invention (2)

Figure 5: Implementation of the video capture step of the present invention (1)

Figure 6: Implementation of the video capture step of the present invention (2)

In order to make the technical means of the present invention and the effects thereof can be more completely and clearly disclosed, the details are as follows. Please refer to the disclosed drawings and drawings: first, please refer to the first and second As shown in the figure, the method for measuring the type of clothes worn by the video device according to the present invention comprises the following steps: A. System calibration: Please refer to the third and fourth figures together, and the model is first (4) Placed in a fixed point (5) located in front of the back plate (6), and a sensing unit (61) is disposed in the middle of the back plate (6) to sense with the video device (1) to assist the video device ( 1) Perform pixel and distance determination, and then correct the distance between the video device (webcam) (1) and the fixed point (5) and the back panel (6), and wear the back panel (6) and the model (4). The figure of the clothes, and the back panel (6) and the model (4) wear the clothes in the opposite color system, the operation unit (2) to capture the image; B. video image: followed by Causing the arithmetic unit (2) to drive the video device (1) to take an image of the model (4) wearing the clothes; C. pixel analysis: then, the arithmetic unit (2) will be worn Models with clothes (4) image pixels for analysis; D. grayscale and binarization: followed by grayscaled and captured images Value processing, and the grayscale image is a two-dimensional matrix image, therefore, only the intensity changes from black to white, without any color information, and generally from red [R], green [G], blue [B Compared with the three-dimensional matrix of the color image formed by the three primary colors, the storage space and the calculation speed of the image processing are much reduced; and the binarization processing is performed, and the binarization processing is to set a threshold [CR], the valve The value is the depth value of the custom color, and then the red [R], green [G], and blue [B] values above the threshold are selected as bright points, and the values below the threshold are dark points, and the calculated values are followed. For processing, the program mode is as follows:

Private Sub Clk( )

If Gray<CR Then Gray=0

If Gray>CR Then Gray=255

End If

From the binarization threshold extraction technique, the coordinates of the correct target position can be found, and the coordinate data can be accurately analyzed after taking the coordinate data; E. Drawing the outline of the clothes line: Continued, the clothes of each model will be selected. [S, M, L, XL, XXL] is worn on the model (4), and then the computing unit (2) drives the video device (1) starting from the center of the model (4), making the lens dependent on the model ( 4) The outer edge of the shoulder (a), the upper arm above the elbow outer edge (b), the elbow to the upper edge of the lower arm (c), the lower arm below the elbow outer edge (d), the elbow to the upper arm The outer edge (e) and the vertical arm (f) of the upper arm to the hem of the garment are sequentially rotated from right to left, from top to bottom, and then left to right to scan the outline of the garment to obtain (a), (b) Straight lines of paragraphs (c), (d), (e) and (f), Then, with the coordinate distance of each straight line, you can know the size of the clothes. After entering the image of the clothes and related sizes into the computing unit (2) database, and then according to the size of each type of clothing; F.撷Take the customer's body contour: please refer to the fifth and sixth figures together. When the customer (7) wants to buy clothes, repeat the above A. system correction, B. video acquisition, C. pixel analysis, D. gray Steps such as gradation and binarization and E. line contour extraction, so that the customer (7) stands on the fixed point (5), and according to the customer (7), the color of the existing clothes is matched with the back panel of the opposite color. (6), and then ingest the customer (7) image, and after the pixel analysis and grayscale and binarization processing, scan the image of the customer (7) size in the same order as the size of the clothes; G. Database comparison: Consequently, the image of the customer (7) and the size of the clothing filed by the database of the computing unit (2) are compared, and the closest and larger than the customer (7) The model is selected as the target; H. Suggested output: the clothing model suitable for the customer (7) will be displayed by the display device (3) Out.

Accordingly, the function of capturing the image by the video device (1) of the present invention compares the size of the selected clothing client (7) with the image of each model size pre-inputted in the database of the computing unit (2). Here, the customer (7) no longer needs to actually try on the clothes when purchasing the clothes, and only through the screening of the system of the invention, the suitable size clothes can be judged according to the size of the customer (7), thus Not only can save customers (7) the time to buy clothes, but also reduce the inconvenience of collecting and sorting clothes.

As can be seen from the above structures and embodiments, the present invention has the following advantages:

1. The present invention consists of a video device (webcam), an arithmetic unit and a display device. The video device is used to capture various types of clothes, and the image of the customer's body shape is purchased, and then the arithmetic unit is used to input the images of the two. The comparison is performed, and then the display device displays the comparison result, thereby simplifying the system design and operation mode, and the effect of quickly screening the appropriate size clothes for the customer can be achieved.

2. The present invention uses grayscale and binarization processing to grayscale color images and then use binarization to determine the correct size of the captured image, thus speeding up image storage and operations. The speed and accuracy of the image comparison.

3. The invention can quickly capture the size of the customer's body shape, and then compare with the size of each type of clothing pre-stored by the computing unit. Here, when the customer purchases the clothes, the customer does not need to actually try on, and the user can be selected to suit the customer. Body clothes to save customers time to buy clothes.

4. The invention can help the customer to select suitable clothes without actually trying on the clothes when purchasing the clothes, thereby saving the trouble for the store to collect and organize the clothes, in addition to saving the time for the customers to purchase the clothes.

In summary, the embodiments of the present invention can achieve the expected functions, and the specific structures disclosed therein have not been seen in similar products, nor have they been disclosed before the application, and have fully complied with the requirements and requirements of the Patent Law. If you apply for an invention patent in accordance with the law, you are welcome to review it and grant a patent.

Claims (9)

A method for measuring a type of clothes worn by a video device, the steps of which include: A. System calibration: first correcting the distance between the video device and the backplane and the target, and correcting the pixels of the photographing backplane and the target. The target is one of a model or a customer wearing clothes; B. video capture: an image of the target device being captured by the video device; C. pixel analysis: and the computing unit is directed to the target object The image is analyzed by the pixel; D. Gray-scale and binarization processing: Continue to make the image of the target object of the captured object first gray-scaled, and then binarized to facilitate the accuracy of the object coordinates. E. Capture the outline of the clothes line: then, drive the video device through the computing unit to input the captured image into the database of the computing unit according to the contour of each type of clothing worn on the model; F. Capture the customer's body shape Contour: The video unit is driven by the computing unit to capture the image according to the contour of the customer; G. Database comparison: comparing the contour of the captured customer's body shape with the outline of the clothing filed by the database; H. Suggestion lose Out: Display the output of the clothing model suitable for the customer. The method for measuring the type of clothes worn by the video device according to the first aspect of the patent application, wherein, in the step of system calibration, the colors of the clothes worn by the backboard and the target are in opposite colors. The method of measuring the type of clothes worn by the video device as described in claim 1 or 2 The method further includes a sensing unit in the middle of the backboard for sensing with the video device. The method for measuring the type of clothes worn by the video device according to the third aspect of the patent application, wherein the front side of the backboard is further provided with a certain point for the target to be placed thereon. The method for measuring the type of clothes worn by the video device according to the first aspect of the patent application, wherein the grayscale system makes the image of the target object obtained into a two-dimensional change only between black and white. Matrix image. The method for measuring a wearing type by a video device according to the first or fifth aspect of the patent application, wherein the binarization processing sets a threshold value, and then selects a red, green, and blue value having a depth greater than a threshold value. For the bright spot, the value below the threshold is a dark point, thereby taking advantage of the coordinates of the correct position of the target. For example, in the method of claim 1, the method for measuring the type of clothes worn by the video device, wherein the outline of the drawn clothes line is starting from the center of the model, and the lens of the video device is based on the outer edge of the model shoulder. The upper part of the upper arm to the outer edge of the elbow, the outer edge of the elbow to the lower arm, the lower arm to the outer edge of the elbow, the outer edge of the elbow to the upper arm and the vertical outer edge of the upper arm to the hem of the clothes are sequentially arranged by the right. To the left, from top to bottom, then reciprocally scan the outline of the clothes from left to right to obtain each line of straight lines, and then the distance of the intersection of each line of straight lines can be used to know the size of the clothes, followed by the clothes and related sizes. The image is input into the database of the computing unit, and then classified according to the size of each type of clothing. The method for measuring the type of clothes worn by the video device according to claim 1 or 7, wherein the order of the body shape of the customer is taken from the center of the customer's body, so that the lens is on the outer edge of the shoulder of the customer. From the top of the arm to the outer edge of the elbow, the outer edge of the elbow to the lower arm, the lower arm to the outer edge of the elbow, the elbow to the outer edge of the upper arm and The vertical outer edge of the upper arm to the hem of the garment is sequentially from right to left, from top to bottom, and then reciprocally scanned from left to right to capture the contour of the customer to obtain the straight lines of each segment, and then the intersection of the straight lines of each segment The coordinate distance will tell you the size of the customer. A method for measuring a type of clothes worn by a video device as described in claim 1 is wherein the video device is a webcam.