JPS63113678A - Lens part recognition system for spectacle frame - Google Patents

Abstract

PURPOSE:To perform easy, high-accuracy recognition by converting a spectacle frame image into binarization and finding information on a frame part, and labeling the background part encircled with the frame part from the center part to the peripheral part and recognizing lens parts in the spectacle frame image. CONSTITUTION:A lens recognition part 8k labels binarized pattern data in the inside area of the frame part indicated by data '1' as a lens part. The lens parts of the spectacle frame are specified by the labeling data with high accuracy. Further, if there is a deficient of the frame part and the labeling of data '2' indicating the lens parts reaches the side part of a rectangular area obtained by cutting the spectacle frame image beyond the original frame part, the labeling result of its periphery, e.g. the labeling result which is one scanning line before is referred to correct the labeling.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to a lens part recognition method for an eyeglass frame that can easily and accurately determine the lens part of the eyeglass frame.

(Prior Art) Recently, images of various eyeglass frames have been recorded, and the images of the eyeglass frames are selectively combined with the image of the face of the person selecting eyeglasses input by the left camera of the TV. One idea is to select a desired eyeglass frame while monitoring a composite image. It has also been considered to combine information about lenses dyed in a desired color with the lens portion of a selected eyeglass frame, and select a lens color and its dyed area suitable for the eyeglass frame from the combined image.

However, when processing such a lens portion, it is necessary to accurately determine the area of the lens portion in the eyeglass frame image. Conventionally, for example, an eyeglass frame image is displayed on a TV monitor, and a person manually traces and instructs the eyeglass frame frame on the displayed image, and the inner area of the frame frame is designated as the lens. I am trying to recognize and process it as .

(Problems to be Solved by the Invention) However, the task of tracing the frame portion of the eyeglass frame image in this manner is extremely troublesome and requires a large amount of processing time. Furthermore, since the tracing of the eyeglass frame is performed manually, there is a lack of accuracy, and there are problems such as the inability to recognize the lens portion with high precision.

The present invention has been made in consideration of these circumstances, and its purpose is to provide an eyeglass frame whose lens portion can be easily and highly accurately determined from an image of the eyeglass frame without any manual effort. An object of the present invention is to provide a lens part recognition method.

[Structure of the Invention] (Means for Solving the Problems) The present invention inputs an eyeglass frame image, binarizes the eyeglass frame image to obtain information on the frame frame, and
The pattern data obtained through digitization processing is sequentially labeled from the center of the background surrounded by the frame to the periphery to recognize the lens portion in the eyeglass frame image.

In particular, taking into account the left-right symmetry of the eyeglass frame, one side of the eyeglass frame in the eyeglass frame image is selectively cut out to perform the above-mentioned labeling process, and the frame indicated by the binarized pattern data is If there is a lack of information in the frame (discontinuity), the labeling process is controlled by referring to the labeling results around the frame, allowing the lens to be recognized simply, efficiently, and with high precision. be.

(Function) Thus, according to the present invention, since it is clear that the background portion surrounded by the frame portion in the eyeglass frame image is the lens portion, the pattern data obtained by binarizing the eyeglass frame image is The lens portion is determined by sequentially labeling the background area surrounded by the frame area from the center position to the periphery. If there is a chip in the frame area due to the accuracy of the binarization process, the labeling result of the surrounding area is Since labeling is controlled by reference, the background region surrounded by the frame can be easily and highly accurately recognized as a lens portion.

(Example) Hereinafter, an example of the present invention will be described with reference to the drawings.

FIG. 1 is a schematic configuration diagram of an eyeglass frame recording device constructed by applying the method of the embodiment, and FIG. 2 is a diagram showing an example of the configuration of a processor used in the device of the embodiment.

This device obtains information on eyeglass frames used in the eyeglass frame selection system, records this information on a predetermined recording device (for example, a floppy disk recording medium), and constructs a database regarding various eyeglass frames. .

In FIG. 1, an eyeglass frame 1 to be recorded is a TV.
An image is inputted by the camera 2, converted into digital data via the A/D converter 3, and temporarily stored in the image memory 4.

This imaging input system is realized using a known digital color image technology, and, for example, obtains an image signal of an input glasses frame as an 8 (bit) RGB signal. These color-separated R, G, and B signals are stored in the image memory 4, respectively. It should be noted that although the explanation will be given here assuming that the eyeglass frame image is obtained by color-separating the RGB signals, it goes without saying that the eyeglass frame image may be obtained as a YIQ signal.

The eyeglass frame image temporarily stored in the image memory 4 is read out via the D/A converter 5 and supplied to the TV monitor 6 for image display. This TV
While checking the eyeglass frame image displayed on the monitor θ, the zooming adjustment of the imaging lens system of the TV left camera and the imaging field of view adjustment are performed, and the TV
Performs cursor control on monitor B.

Through such adjustment, an image of the eyeglass frame is obtained in a desired size, and the eyeglass frame image of a predetermined rectangular area containing the image of the eyeglass frame is stored in the image memory 4 as an image to be processed by the processor 8, which will be described later. . For example, the third
As shown in figure (a), 200 pixels in the horizontal direction, pixel surface direction 1
The image of the eyeglass frame should fit within the rectangular area of the 00 line, or the image of one side of the eyeglass frame should fit within the rectangular area of 100 pixels in the horizontal direction and 100 lines in the vertical direction, as shown in Figure (b). At this time, a shutter signal is applied from the control unit 7 to store and hold the captured image data of the rectangular area in the image memory 4.

After cutting out the image of the eyeglass frame in this way, the processor 8 performs image processing to recognize and extract the frame frame portion, and generates pattern data (identification data) that identifies the frame frame portion and other background portions. generate. Specifically, for the rectangular area, a binary pattern data is obtained in which, for example, the frame portion of the glasses is set to "1" and the background part is set to "O".

Here, assuming that the processor 8 performs image processing only on one half (right side) of the eyeglass frame, first the RAM
Image data (RGB signals) of a rectangular area of 100 pixels in the horizontal direction and 100 lines in the vertical direction are stored in the image memory 4a.
Read from and store. In this way, if you extract only one half of the eyeglass frame image from the eyeglass frame image,
The processing amount can be cut in half.

After that, from the image data stored in this RAM 8a,
As shown in FIG. 3(b), image data (RGB signals) at the center position of the rectangular area is extracted as background image data and stored in the RAM 8b.

The extraction of this background image data is as follows: - Normally, when imaging an eyeglass frame, the eyeglass frame is set in front of a predetermined back screen, and the background image data is approximately uniform; It supports the lens on the inside, and as mentioned above, when only one half of the eyeglass frame is captured, the center part is usually the part into which the lens is fitted (lens part), that is, the background part to the frame frame. It is specified based on the viewpoint. By identifying the center position of the rectangular area as the background part in this way, it is possible to almost certainly extract the image data of the background part, regardless of the shape and size of the eyeglass frame frame, or regardless of slight positional deviation. can.

The arithmetic unit 8C scans and sequentially reads out the image data (RGB signals) stored in the RAM 8a.
The calculation H=lR-rl+lG-gl+IB-bl is performed between the foreground image data (rgb signal) stored in 8b, and the discrimination function value H of the frame frame is determined. This discrimination function value H is normally (R, r
), CG, g).

The numerical value is shown to the extent that at least one of (B, b) is significantly different. Conversely, if the image data of the pixel read out from the RAM 8a is for the background part, (R, r
), (G, g), and (B, b) are almost equal, indicating "0" or a value close to it.

The binarization circuit 8d binarizes the discrimination function value H using a predetermined threshold value to obtain a signal that is "1" in the frame portion and "0" in the background portion. The binary signals (pattern data) thus obtained are sequentially stored in the memory 8r under the control of the address circuit 8e in synchronization with the reading of image data from the RAM 8a. As a result,
Pattern data indicating the frame portion in the rectangular area is stored as binary image data in the memory 8r.

Here, the lens recognition unit 8 is configured to label the internal area of the frame portion indicated by data "1" as a lens portion with respect to the pattern data subjected to the binary processing as described above. Specifically, as described above, since the center position of the rectangular area can be identified as the lens part located inside the frame portion, the pixel at the center position is labeled as pattern data "2".

Then, the pixel of interest is scanned from the center coordinates ('x, y) to the right (left), and if the data of the pixel of interest is "0", it is labeled as "2". The process is repeated until pattern data "1" indicating a frame frame is detected.Then, at the time when a frame frame is detected, the pixel of interest is returned to the center pixel position, and scanning in the opposite direction is performed in the same manner.

After that, by changing the scanning line and repeating the same process, the internal area surrounded by the frame portion indicated by the pattern data "1 degree" becomes the data "
2'', and the lens portion of the eyeglass frame can be specified with high precision by this labeling datum.

However, depending on the type of eyeglass frame, some have thin frames, and some have a similar color to the background. In such a case, a part of the frame frame may not be detected as a frame frame portion indicated by pattern data "1" due to the binarization process described above. In other words,
In some cases, gaps (discontinuities) occur in the data ladder 1" indicating the frame frame portion in the binarized pattern data.

If there is such a lack of information in the frame frame area, when labeling is performed by scanning from the center coordinates to the peripheral area as described above, the frame frame area indicated by data "1" will not be detected, so the scanning will be incorrect. The frame of the rectangular region cut out from the eyeglass frame image is reached, and undesirable labeling is performed.

Therefore, in this method, as shown in FIGS. 4 and 5, for example, if there is a chip in the frame portion, the labeling of data “2” indicating the lens portion may extend beyond the original frame portion in the eyeglass frame image. When reaching a side of a rectangular area cut out, the labeling is corrected by referring to the labeling results around the edge, for example, the labeling results from one scanning line before. Specifically, as shown in FIGS. 4 and 5, data "1" is labeled in the area where the frame frame part is missing based on the connection relationship of pixels assigned data "1", and the data "1" is labeled according to this data. 2” is used to control the pixels to be labeled. Such labeling control prevents an area beyond the original frame portion from being labeled as a lens area.

Note that when performing such labeling processing, for example, in the case of a spectacle frame as shown in FIG. 6, it becomes impossible to label the area shown in the hatched area in the figure. In other words, when the first pixel to be scanned indicates a frame frame portion, the labeling process for that scanning line ends at that point.

In such a case, as shown in Figure 7, the surrounding 8
What is necessary is to check the pixels (3×3 pixel area) and shift the scanning start point. With this kind of labeling control,
It becomes possible to effectively label the entire internal area surrounded by the frame portion as a lens portion.

Thereafter, the pattern data written in the memory 8r is updated with the thus labeled data. As a result, pattern data consisting of data "1" indicating the eyeglass frame portion, data "2" indicating the lens portion, and data "O" indicating the other background portion is stored in the memory 8r.

The memory 8r for storing pattern data having three values may be any memory having a capacity of 2 bits per pixel.

Then, the pattern data stored in this memory 8f,
Sequential readout and synthesis circuit 8 under the control of address circuit 8o
g to the recording device 9 for recording.

At this time, when the pattern data read from the memory 8f and output to the recording device 9 takes a value of "1" indicating the frame frame portion, the synthesis circuit 8g reads the image data (RGB signals) of the corresponding portion from the RAM 8a. , this is synthesized and output. As a result, along with the scanning of the pattern data, only the image data (RGB signals) of the frame portion are sequentially associated with the pattern data "1°" and output, and are recorded in the recording device 9 together with the pattern data. .

Note that the information on the spectacle frames recorded in this manner is reproduced as follows and restored to the original image data.

That is, the information reproduced from the recording device 9 is input to the separation circuit 8h, and is separated into image data consisting of pattern data and RGB signals. The separated pattern data are sequentially stored in the memory 8r, and the pattern data "
The image data associated with pattern data "1" is separated from the pattern data "1" and sequentially stored in another memory 81.

The address circuit 8e sequentially scans and reads out the pattern data stored in the memory 8r, and selectively controls the switch 8j according to the pattern data. Then, when the pattern data is "0°", the fixedly given image data "0;black" is selected, and when the pattern data is "1", the image data stored in the memory 8I is sequentially read out. and is written to the RAM 8a. Furthermore, if the pattern data is "2", select the color information (RGB signal) input from the outside,
This is written into the RAM 8a.

Through this process, the image data obtained from the memory 81 is sequentially written into the frame frame portion indicated by the pattern data “1” in the RAM 8a, and the image data obtained from the memory 81 is sequentially written into the frame frame portion indicated by the pattern data “0”. Black image data "0" is sequentially written into , and the above-mentioned image of the eyeglass frame is restored.

Furthermore, image data consisting of designated color information is written into the lens portion inside the frame indicated by pattern data "2".

By transferring the image of the glasses frame restored to the RAM 8a to the image memory 4, an image of the glasses frame consisting of the glasses frame and lenses colored in the designated color is displayed on the TV monitor 6. The image will be displayed.

At this time, if only the information on one half of the eyeglass frame is recorded as described above, the image on the other side is generated by processing the eyeglass frame image of the one half half reproduced and restored in the RAMga. .

Then, these two images (the right half image and the left half image) may be combined to form one eyeglass frame image, and this may be written into the image memory 4.

As explained above, according to the present invention, an input eyeglass frame image is binarized, and the binarized pattern data is sequentially labeled as a lens part from the inner area surrounded by the frame frame to the periphery. Therefore, the internal area surrounded by the frame can be easily and accurately recognized as a lens portion. Moreover, since the labeling of the chipped portion of the frame is controlled by referring to the labeling results of the surrounding area, the lens portion can be effectively resized without being affected by errors in the binarized pattern data. Can be labeled.

Therefore, in accordance with the labeling results, it is possible to carry out processing such as coloring the lens portion in a desired color very simply and accurately, and other great effects can be achieved.

Note that the present invention is not limited to the embodiments described above. For example, the direction of labeling scan from the center to the periphery inside the frame and its scan control can be modified in various ways. Furthermore, various interpolation methods can be applied to pattern data as a method for correcting missing portions of the frame. In addition, the present invention can be implemented with various modifications without departing from the gist thereof.

[Effects of the Invention] As explained above, according to the present invention, 1. the lens of the eyeglass frame is processed by labeling the eyeglass frame image with binarized pattern data without any manual effort such as tracing the eyeglass frame image; parts automatically,
Moreover, effects such as being able to recognize with high accuracy can be achieved.

[Brief explanation of the drawing]

The figures show one embodiment of the present invention, and FIG.
The figure shows an example of the configuration of the processor in the apparatus of the embodiment, and FIGS. 3 to 7 are diagrams for explaining a processing procedure for recognizing a lens part in an eyeglass frame. 1... Glasses frame, 2... TV left camera 4...
Image memory, 6... TV monitor, 7... Control section, 8... Processor, 9... Recording device, 8a,
8b...RAM. 8C... Arithmetic unit, 8d... Binarization circuit, 8f...
Memory, 8g... synthesis circuit, 8k... lens recognition unit (labeling processing). Applicant's agent Patent attorney Takehiko Suzue Figure 2 (Q) Figure 3 Figure 4 Figure 16 (b) Figure 5 Figure 7

Claims (4)

[Claims] (1) means for inputting an eyeglass frame image; means for binarizing the input eyeglass frame image to obtain information on the frame frame; A method for recognizing a lens part of an eyeglass frame, comprising means for recognizing a lens part in the eyeglass frame image by sequentially labeling a background part surrounded by parts from a central position to a peripheral part. (2) The method for recognizing the lens portion of an eyeglass frame according to claim 1, wherein the binarization processing and the labeling processing are performed by selectively cutting out one side of the eyeglass frame in the eyeglass frame image. (3) The eyeglass frame lens portion recognition method according to claim 1, wherein the eyeglass frame image is input as a color image. (4) If there is a lack of information in the frame portion indicated by the binarized pattern data, the labeling process is controlled by referring to the labeling results around it. Lens part recognition method for eyeglass frames described.