JPH08101068A - Color measuring apparatus - Google Patents

Abstract

PURPOSE: To obtain a color measuring apparatus for measuring the true color of an object conveniently with high accuracy regardless of the number and position of illuminators or the image pickup position. CONSTITUTION: The color measuring apparatus comprises a coordinate conversion circuit for receiving analog R, G, B coordinate system signals influenced by illumination from a camera 1 and converting them into a brightness (lightness) color component and a color component insusceptible to the brightness (hue and color saturation), a signal processing section 200 for statistically processing the converted color components to analyze the density frequency, and an output section 300 for displaying or printing an output signal from the signal processing section 200.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color measuring device for an object, and more particularly to a color measuring device for an object having a non-uniform shape.

[0002]

2. Description of the Related Art An object is photographed using a television camera or the like,
Measurement and evaluation of the color have been performed conventionally. For example, Japanese Examined Patent Publication No. 61-7360 discloses a technique in which Prince Melon is photographed by an ITV camera, the color and pattern of the fruit are measured based on the image signal, and the grade of the fruit is determined.

In addition, the color is also measured by using a spectrophotometric method. The spectrocolorimetric method is a method of obtaining a spectral reflectance characteristic at a measurement point, performing processing according to human color perception, and calculating a color.

[0004]

However, in order to measure the color of an object using an image signal from a camera or the like, it is very difficult to set the conditions for illumination and photographing. That is, the image signals obtained from the camera are usually red (R), green (G), and blue (B) signals, but these color information have brightness (= illumination).
Contains the ingredients of. This can be understood from the fact that, for example, when a red sphere is photographed from above, it is photographed as white → bright red → dark red → black from a portion close to the illumination. Furthermore, it is still good if the shape of the measured object is uniform, but if the shape of the measured object is not uniform, the measurement becomes more difficult due to the effect of shadows, and even if the object is actually the same color, the measurement results will vary. It becomes a thing.

On the other hand, in the spectrophotometric method, it is difficult to measure a wide range of colors and color spots because the amount of measurement data at one point is large and the range that can be measured at one time is very small.

[0006]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a color measuring device capable of accurately and simply measuring the true color of an object regardless of the number and position of illuminations and the photographing position. For this reason, in the present invention, the brightness (brightness) component and the color component not affected by the brightness (from the signal of the R, G, B coordinate system including the influence of illumination, which is obtained from the image pickup means such as a camera ( The objective is achieved by providing a signal processing unit that obtains the (hue and saturation) component and a statistical processing unit that detects the color density frequency distribution and color spots of the measured object from the output signal of the signal processing unit. are doing.

Here, the operating principle of the present application will be described. As described above, the R, G, B signals obtained by the camera or the like include the influence of illumination. Therefore, this signal is separated into a brightness component and another component, that is, a component representing the original color of the measured object, and color measurement is performed based on this component representing the color. Specifically, the separation is performed based on the following formula.

In each of the following equations, R, G, and B represent values obtained by normalizing the digitally converted R, G, and B signal values in the range of 0 to 1, respectively. Also, the function min (R, G,
B) is the minimum value of R, G, B values, and similarly, mid (R,
G, B) represents an intermediate value, and max (R, G, B) represents a maximum value. The converted HSL coordinate system is

[0009]

[Equation 2]

By this conversion, the R, G, B orthogonal coordinate system (FIG. 3A) is standardized to the HSL coordinate system (FIG. 3A).
The coordinates are converted into (b)). As can be seen from FIG. 3B, in the HSL coordinate system after conversion, the lightness (L) representing the brightness is a value on the center line of the cylinder, and therefore the color indicated by the displacement in the radial direction from the center line. The degree (S) is independent of the hue (H) indicated by the circumferential angle of a circle whose origin is the center line. Therefore, by measuring the color of the object to be measured using the values of saturation and hue excluding lightness, it is possible to obtain the same color substance regardless of the lighting conditions or the shape. The same measurement result can be obtained.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail with reference to the drawings. FIG. 1 is a schematic diagram showing the configuration of an embodiment of the color measuring device of the present invention. The color measuring device includes an imaging unit 100,
The signal processing unit 200 and the display unit 300 are included. The image pickup unit 100 supports a camera 1 using an image pickup device such as a CCD, and supports the camera 1 and the camera 1
The cooling case 11 that promotes heat dissipation from the camera, the camera support 12 that holds the cooling case 11 and supports the imaging surface of the camera in parallel with the measurement table 13 at a predetermined distance, and the xenon lamp, etc. And a lighting means 5 for illuminating the measuring table 13 with a light guide means such as an optical fiber.

The signal processing section 200 receives the analog signal from the camera 1, performs various signal processing as described later, and outputs the result to an output section 300 composed of a CRT and / or a printer. Next, the operation will be described with reference to FIG. FIG. 2 is a block diagram of the signal processing unit 200 in FIG.

The camera 1 is, for example, R (red), G (green),
It has three CCDs each corresponding to B (blue) and having a predetermined number of pixels.
The analog signal of is independently generated. These signals are
The image is input to the image input unit 210 of the signal processing unit 200. R,
The G and B signals are respectively input to the image input unit 210 from the first to the first.
Third analog / digital (A / D) conversion circuit 211 to
By 213, for example, each pixel is converted into a digital value of 256 steps. The digital values are written in the first to third frame memories 214 to 216, respectively,
The signal processing circuit 220 performs signal processing such as coordinate conversion.

The coordinate conversion circuit 221 has R, G, and
Based on the digital signal of each color B, it is converted into a signal of H, S, L coordinate system. After the coordinate conversion, the H, S, and L values are written in the fourth to sixth frame memories 222 to 224 provided for the H, S, and L signals, similarly to the digital R, G, and B values. Next, the statistical processing unit 225 reads the hue (H) data from the fourth frame memory 222 and the saturation (S) data from the fifth frame memory 223, and analyzes the density frequency distribution. Assuming that the concentration frequency distribution is a superposition of normal distributions,

[0015]

(Equation 3)

The mean value (μ) and the variance (σ ² ) are determined by multiple regression calculation using the normal distribution function represented by
Also, the standard deviation (σ) can be obtained from the variance. The average value indicates the color component of the imaging surface, and the dispersion indicates the degree of color spots. These analysis results are displayed and / or printed on the output unit 300 as needed. These series of operations are controlled by a control device (not shown). A commercially available general-purpose microcomputer can be used as the control device.

Next, application examples using this apparatus will be described. Tomato was used as the object to be measured, and changes in the measurement results depending on the ripeness thereof were examined. A / D conversion was performed in 256 steps for each of R, G, and B per pixel, and diffused light from a xenon lamp (color temperature 5500K) was used for illumination. The lamp was set vertically with respect to the measuring table so as to have regular reflection on the measuring table. Also, in order to ensure accurate measurement, the offset output of the camera at an illuminance of 0 (dark) is measured,
A process of removing this offset from the color measurement value was performed.

The unripe tomatoes were stored in a thermostatic chamber at 10 ° C., the surface color was measured every day during the ripening process, and the change with time was observed. As immature tomatoes, tomatoes that started to turn slightly red (Experiment 1) and almost green tomatoes (Experiment 2) were used. A tomato image was obtained with a black background, and the completely reflected portion was removed from the obtained image. The removal of the completely reflected portion was performed by utilizing the color difference and regarding the portion having a color difference to some extent as compared with white as the completely reflected portion.

Next, extract the tomato image from the background,
The analog RGB signal was digitized for each pixel and further converted into an HSL signal. The average value and standard deviation of the hue (H) and the saturation (S) were obtained. FIG. 4 shows the standard deviation of hue, FIG. 5 shows the average value of hue, FIG. 6 shows the standard deviation of saturation, and FIG. 7 shows the change over time of the average value of saturation.

In FIG. 4, within about 10 days from the start of observation, variations in standard deviation are recognized due to the difference in ripening condition at the start of measurement. Especially in Experiment 2 in which almost green tomatoes were used, the standard deviation was low at the beginning due to the relatively uniform green color, but as the coloration began to become red, color unevenness occurred and the standard deviation increased. However, when the fruit became ripe, it turned into a substantially uniform red color, and converged to a standard deviation almost the same as in Experiment 1.

Further, it can be seen from FIG. 5 that although the hues at the start of the experiment are different, they finally converge to almost the same hue. Further, although there are some individual differences in the saturation in FIGS. 6 and 7, it can be seen that almost the same changes are exhibited.

[0022]

As described above, according to the present invention, it is possible to obtain substantially the same color measurement results for objects having different shapes. Therefore, for example, the hue and saturation of a fruit having a preferable ripeness can be obtained in advance. By measuring with the device of the present invention and performing the selection based on the measured value, it is possible to realize the selection with extremely high precision as compared with the conventional similar device.

[Brief description of drawings]

FIG. 1 is a diagram showing a configuration example of a color measuring device according to the present invention.

FIG. 2 is a block diagram of a signal processing unit in FIG.

3A and 3B are diagrams showing display coordinates of an image signal.

FIG. 4 is a diagram showing a change with time of a standard deviation of hue in an application example.

FIG. 5 is a diagram showing a change with time of an average value of hues in an application example.

FIG. 6 is a diagram showing a change with time of a standard deviation of saturation in an application example.

FIG. 7 is a diagram showing a change with time of an average value of saturation in an application example.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 camera 5 illumination means 11 cooling case 12 camera support stand 13 measurement stand 200 signal processing section 300 display section

Claims (2)

[Claims] 1. A television camera means for photographing an object to be measured from a predetermined direction and outputting an image signal thereof as an analog RGB signal, an illuminating means for illuminating the object to be measured from a predetermined direction, and the analog RGB signal. A signal converting means for receiving and converting the analog signal into a hue (H) signal, a saturation (S) signal, and a brightness (L) signal, and an operation for obtaining and outputting an average value and a variance of the hue signal and the saturation signal. A color measuring device comprising: a means and an output means for displaying or printing an average value and a variance of the hue signal and the saturation signal. 2. The analog RG, wherein the signal conversion means is
An analog / digital conversion means for converting the B signal into a digital signal with a predetermined accuracy and outputting it as a digital RGB signal, and a conversion operation section for converting the digital RGB signal into the hue signal, the saturation signal and the lightness signal according to the following formula. The color measuring device according to claim 1, wherein the color measuring device comprises: [Equation 1] However, in each equation, R, G, and B represent values obtained by normalizing the digitally converted R, G, and B signal values in the range of 0 to 1, respectively. Also, the function min (R, G, B) is R,
Similarly, mid (R, G, B) represents an intermediate value, and max (R, G, B) represents a maximum value of G and B values.