JPS6215421A - Color recognizing device for object - Google Patents

Abstract

PURPOSE:To curtail the capacity of a memory, and to execute a high speed image processing by discriminating a specified color by a combination of an address signal which has been obtained in a data terminal and an address signal corresponding to the remaining video signal of video signals of three colors. CONSTITUTION:Each density value G, B of green and blue is converted to digital signals GD, BD by analog-to-digital converters 14G, 14B, and inputted to a memory 16 as address signals. A digital signal BG from a data terminal D16 of the memory 16 and a digital signal RD from an analog-to-digital converter 14R are inputted to a memory 17 as address signals, respectively. In this way, by a combustion of these address signals, a binary data which has discriminated a specified color is outputted. In this way, the capacity of the memory can be curtailed, comparing with a conventional one.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to a color recognition device for an object, and in particular to an apparatus for recognizing the color of an object using color as a medium for online inspection of defects, shapes, etc. of a measurement object that has been specified to a certain extent. This invention relates to an object color recognition device used.

<Prior Art> The color recognition device shown in FIG. 5 is a conventional color recognition device that is the basis of the improvement of the present invention. Reference numeral 10 denotes a lens for inputting and condensing a video signal, and dichroic mirrors 11 and 12 convert the condensed incident light into three colors of red, green, and blue (R, G).

B), and each color is divided into image sensors 13R and 13G.
.

13B. Image sensor 13R, 13G, 13
B converts it into an analog electrical signal corresponding to the density value of each color, and converts this into an analog-digital converter 14R, 14G.
.

14B. Analog to digital converter 14R2
14G and 14B convert this electric signal into a digital signal RD.

The signal is converted into GD and BD and input to the memory 15 as an address signal. The memory 15 outputs a binary image output PD corresponding to a predetermined combination of input address signals to its data terminal.

By the way, FIG. 6 shows the density values R,
G and B are decomposed and expressed into three dimensions to show a nine-color vector space. A color image taken of the object to be measured is a pixel Q (R, G, B) whose density values of red, green, and blue signals are R, G, and B.
It is expressed as a set of , but if you change R, G, and B at the same rate in a certain color part of the image, the hue will not change, only the brightness will change. is shown as The part showing a specific color in an image showing similar colors, that is, the range indicated by the trapezoidal cone γ in Fig. 6 is B.
, G, the range is shown by diagonal lines in the figure. When converting this range into a binary image, if (BIG) is in the shaded area shown in FIG. If you write binary data so that it is output to the data terminal, and give the digital signals GD and BD as addresses in the memory 15, a binary image will be generated that outputs 111 only when the area has a specific color. Obtained within the period. The above points are exactly the same in the case of three-dimensional display.

<Problems to be Solved by the Invention> However, in such a conventional color recognition device, the memory size is limited to one screen of a binary image, assuming that the data length of each RGB digital signal is n bits. Hit z3n
Becomes a bit. If P is the number of screens of a binary image, that is, the number of areas to be detected, the total memory amount is 231 × p
becomes. If n = P = -8, the total memory amount is 1
It becomes extremely large at 28M bits. Although the price of memory is decreasing, the amount of 128 Mbit still accounts for most of the device cost. Furthermore, it is necessary to write data for distinguishing between color recognition areas in the memory, but if the memory is large, there is a problem in that it takes a lot of time to write the data.

<Means for Solving the Problems> In order to solve the above problems, the present invention provides means for obtaining three-color video signals of red, green, and blue related to the colors of an object to be measured, and a method for obtaining video signals of three colors related to the colors of an object to be measured. An analog/digital converter converts each density value into a digital signal, and two digital signals from the digital left signal corresponding to the three color video signals are input as address signals, and the combination of these address signals determines the measurement target. A first memory element in which a condition for outputting a second address signal by distinguishing a specific color of an object is written, an address signal obtained at a data terminal of this first memory element, and the remaining video of the three-color video signal. The second memory element is provided with an address signal corresponding to the input signal and a second memory element in which conditions for distinguishing a specific color and outputting binary data based on a combination of these address signals are written.

<Example> Embodiments of the present invention will be described below with reference to the drawings.

Note that parts having the same functions as those in the prior art are denoted by the same reference numerals, and description thereof will be omitted as appropriate.

FIG. 1 is a block diagram showing an embodiment of the present invention.

The green and blue density values G and B are converted into digital signals GD and BD by analog-to-digital converters 14G and 14B, and are input to the memory 16 as address signals.

In the memory 16, for example, the combination of address signals corresponding to the specific color to be detected indicated by diagonal lines in FIG. 2, which is a projection onto planes (B, G) in FIG. The digital signal BD is output from the data terminal D16, and data is written so that 10' is output for combinations of address signals outside the diagonally shaded area. Memo January 7th is Memo January 6th
The digital signal BG from the data terminal D16 and the digital signal RD from the analog-to-digital converter 14R are input as address signals, respectively. The memory 17 stores in advance, for example, a combination of digital signals BG and RD corresponding to a portion β of a specific color to be detected shown by diagonal lines in FIG. II
Data is written so that I is output from the data terminal D17, and 101 is output for combinations of address signals outside the diagonally shaded area.

With the above configuration, the data terminals of the memory 17,
The projection onto the (B, G) plane and the projection onto the (BIR) plane are combined, and as a result, the pixel portion corresponding to the part of the trapezoidal pyramid γ shown in Fig. 4 outputs f, 11, and the other In the case of a region, a binary image that outputs the IOI can be obtained, and almost the same function as the case shown in FIG. 5 is exhibited.

However, if the data length of the digital signals BD and GD (7) is each composed of n bits, and the data length of the digital signal BG obtained from the data terminal D16 of the υ memory 16 is also n bits, then the memo n υ amount of the memory 16 is 2 Xn bits. Further, if the number of screens obtained at the data terminal D1□ of the memory 17 is P, the amount of memory in the memory 17n is 2 XP bits. Therefore, their total memory amount is 22 n (n + P) bits. As in the case of FIG. 5, if n = P = 8, 2'' (n + P) = IM bit, and the total memory amount will be about 1/100 compared to the case shown in FIG.

<Effects of the Invention> As described above in detail along with the embodiments, according to the present invention, the memory capacity can be significantly reduced compared to the conventional art, thereby enabling inexpensive high-speed image processing. Furthermore, initialization of the circuit can also be executed in a short time.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing one embodiment of the present invention, and FIG.
FIG. 3 is an explanatory diagram explaining the operation of the first memory in FIG. 1; FIG. 3 is an explanatory diagram explaining the operation of the second memory in FIG.
FIG. 4 is an explanatory diagram explaining the operation when the first and second memories in FIG. It is an explanatory diagram to explain. 10...Lens, 11, 12...Dichroic mirror, 1
3R, 13G. 13B...Image sensor, 14R, 14G, 14B.
・Analog-to-digital converter, 15.16.17...
・Memory, RD. GD, BD, BG...digital signal, PD...image output.

Claims (1)

[Claims] means for obtaining three color video signals of red, green and blue related to the color of the object to be measured; an analog-to-digital converter for converting each density value of the video signal into a digital signal; and said three color video signals. Two of the digital signals corresponding to the first address signal are input as address signals, and a condition is written for distinguishing a specific color of the object to be measured by a combination of these address signals and outputting a second address signal. A memory element, and an address signal obtained at the data terminal of the first memory element and an address signal corresponding to the remaining video signals of the three colors are input, and the specific color is determined by a combination of these address signals. distinguish 2
1. A color recognition device for an object, comprising: a second memory element in which conditions for outputting value data are written.