JP2001074663A - Visual examination evaluation device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To accurately distinguish the ground of an outer layer from the other fringe parts by applying near-infrared rays to the outer layer of a body to be inspected and detecting near-infrared rays at a water absorption band out of reflection light from the surface of the body to be inspected. SOLUTION: Near-infrared rays are applied to a melon (a) with a fringe on a surface layer of, for example the net-family melon, from a light source 1. A filter 2 for passing only the near-infrared rays at a water absorption band out of reflection light from the melon (a) is arranged at the oblique upper portion of the melon (a). An image-inputting device 3 that is composed by an image pickup element being sensitive to light through the filter 2 is arranged at the rear of the filter 2. A computer 4 for processing an image inputs an image being picked up by the image-inputting device 3. Then, a threshold for extracting the fringe of the surface layer of the melon (a), that for extracting the ground of the surface layer, and that for extracting a crack are set, multi-level processing is made based on each value, thus extracting the ground of the surface layer of the melon (a), the fringe of the surface layer, and the crack of a vine and the surface layer.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an appearance evaluation apparatus for evaluating the appearance of a net-type melon having stripes (nets) on the surface of the epidermis.

[0002]

2. Description of the Related Art Conventionally, the appearance of an object to be inspected, such as a net-type melon, having an uneven surface is evaluated by photographing the melon with a color camera and based on the color information of the color image. This was done by identifying and evaluating

[0003]

However, when the distinction between the ground portion of the epidermis and the striped portion is made based on color information, it is susceptible to differences in the degree of coloring and uneven lighting caused by individual differences. However, there is a disadvantage that the identification accuracy is low.

Therefore, the present invention focuses on the fact that there is a difference in the content of moisture and chlorophyll between the ground and stripes of the epidermis in a test object such as a melon, and distinguishes between the ground and stripes of the epidermis with high accuracy. The purpose is to do.

[0005]

In order to achieve the above object, the present invention is configured as follows. That is, the invention according to claim 1 includes a light source that irradiates near-infrared rays toward the epidermis of an object to be inspected, and a detecting unit that detects near-infrared rays in a moisture absorption band from reflected light from the object to be inspected. And an identification unit for identifying a ground surface of the test object and a striped portion other than the ground based on a detection value of the detection unit.

According to a second aspect of the present invention, there is provided a light source for irradiating visible light toward an epidermis of a test object, and detecting means for detecting visible light in a chlorophyll absorption band from reflected light from the test object. And an identification unit for identifying a ground surface of the skin of the object to be inspected and a stripe portion other than the ground based on a detection value of the detection unit.

[0007]

According to the first aspect of the present invention, the detecting means detects near infrared rays in a moisture absorption band from the reflected light from the test object. The detection value of this detecting means is remarkably different between the ground portion of the epidermis containing a large amount of water and the striped portion of the epidermis containing a small amount of moisture. Then, the identification means identifies the ground and the stripe of the epidermis of the test object based on the detected value.

As described above, the distribution of moisture in the epidermis of the test object is detected using near infrared rays in the moisture absorption band, and the ground and stripes in the epidermis are identified based on the detection result. And the identification accuracy is improved. In the invention according to claim 2, the detecting means detects visible light in the chlorophyll absorption band from the reflected light from the test object. The detection value of this detection means is significantly different between a ground portion containing a large amount of chlorophyll and a stripe portion having a small amount of chlorophyll. Then, the identification means identifies the ground and the stripes of the epidermis of the melon based on the detected value.

As described above, the chlorophyll distribution of the epidermis of the test object is detected using visible light in the chlorophyll absorption band, and the ground and the stripes of the epidermis are distinguished based on the detection result. And the identification accuracy is improved.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described. Prior to the description, the principle of the present invention will be described. The invention according to claim 1 focuses on the fact that, for example, a test object having a stripe having a property different from that of the ground on the skin, such as a net-type melon, has a different moisture content between the ground and the stripe of the skin. It was done. Therefore, the results of detecting the reflectance by irradiating near-infrared rays with continuously changing wavelength to each of the ground part of the skin, the striped part, and the vine part of the net-based melon Arls Melon, A reflection spectral characteristic (spectrum) as shown in FIG.

Therefore, the invention according to claim 1 uses the near-infrared ray having a wavelength of about 1.4 μm or about 1.9 μm, which is a water absorption band, based on the reflection spectral characteristics to evaluate the appearance of the object to be inspected. Is what you do. Next, the invention according to claim 2 is that, for example, a test object having a stripe having a property different from the ground on the epidermis, such as a net-type melon, has a different chlorophyll content between the ground and the stripe portion of the epidermis. It pays attention to. Then, as a result of detecting the reflectance by irradiating visible light with continuously changing wavelength to each of the ground portion of the skin of the net type melon and the stripe portion, the reflection spectrum as shown in FIG. 2 was obtained. Characteristics (spectrum) were obtained.

Therefore, according to the second aspect of the present invention, the wavelength, which is the absorption band of chlorophyll, is 0.1 mm based on the reflection spectral characteristics.
Around 67μm or around 0.54μm and 0.67μm
The visible light in the vicinity is used to evaluate the appearance of the test object. Next, an embodiment of the invention described in claim 1 based on the above principle will be described with reference to the drawings.

In FIG. 3, reference numeral 1 denotes a light source for irradiating near-infrared light to a melon a having a stripe on its skin such as a net-type melon. Obliquely above melon a, the water absorption band of the reflected light from melon a (as shown in FIG. 1,
The filter 2 is arranged to pass only near-infrared rays (around 1.4 μm or around 1.9 μm). An image input device (camera) 3 composed of an image sensor (sensor) sensitive to light passing through the filter 2 is arranged behind the filter 2.

The image input device 3 is connected to the input side of the computer 4 for image processing. Image processing computer 4
Processes the input image from the image input device 3 as described later. An output side of the image processing computer 4 is connected to an image output device 5 such as a display device or a printer.

Next, an example of image processing according to the embodiment of the present invention will be described with reference to the flowchart of FIG. Here, a case will be described in which the melon a to be processed is a round melon. Now, the near-infrared ray emitted from the light source 1 is applied to the melon a, and only the near-infrared ray in the moisture absorption band (about 1.4 μm or about 1.9 μm) of the reflected light from the melon a passes through the filter 2. I do. Therefore, an image obtained by photographing with the image input device 3 is formed from moisture information.

Then, the image photographed by the image input device 3 is inputted (S1). Next, setting of a threshold value for extracting the stripes of the skin of melon a (S2), setting of a threshold value for extracting the ground of the skin of melon a (S3), Setting of threshold for extraction (S
4) is performed.

Here, the change in luminance in the vertical direction (vertical direction) based on the output from the image input device 3 is as shown in FIG. Also, the change in luminance in the horizontal direction (horizontal direction)
As shown in FIG. As can be seen from FIG. 5, the level of the vine is lower than that of the detection of the epidermis as a whole, and the level of the ground and the stripes are distinctly different when the epidermis is detected. Therefore, each threshold value for the above extraction is determined in consideration of these points.

Subsequently, multivalue processing is performed based on each of the threshold values (S5), and the ground of the skin of melon a, stripes of the skin,
Extract the cracks of the vine and epidermis. Thus, claim 1
In the embodiment of the invention described in the above, from the information of the water distribution in the epidermis of melon a, the ground of the epidermis of melon a, stripes of the epidermis, vines, and each of the extraction processing of cracks in the epidermis are performed, so that the extraction is performed Can be performed with high precision.

Next, an example of a method for evaluating a crane of a crane with a crane using the apparatus shown in FIG. 3 will be described with reference to FIGS. First, an image captured by the image input device 3 is fetched (S11), binarized by a predetermined threshold to extract a crane portion (S12), and an image as shown in FIG. 8 is obtained. Next, the feret diameter x and the feret diameter y are measured from the image (S13), and the feret diameter x or the feret diameter y out of the range of the following equations (1) and (2) are out of specification. (S14, S15).

X1.ltoreq.x <x2 (1) y1.ltoreq.y <y2 (2) Further, when the constituent pixels in the y direction corresponding to each pixel position are measured from the image of the vine (S16), the result is as shown in FIG. . Therefore, as shown in FIG. 9, the maximum number of pixels do, the number of pixels d1 at a position separated from the maximum number of pixels by a predetermined set value δ,
d2 is measured (S17). Then, the quality of the crane is determined from the obtained values according to the following equation (3). If the determination is negative, the crane is out of the standard, and if the determination is affirmative, it is determined to be good (S18).

Z1 ≦ {do− (d1 + d2) / 2} <z2 (3) Here, z1 and z2 are constants set in advance. By such processing, the evaluation of the crane of the melon with the crane can be performed with high accuracy.

Next, an embodiment of the present invention will be described with reference to the drawings. In FIG. 10, 11
Is a light source for irradiating visible light to the melon a having a stripe on the skin like a net-type melon. A filter 12 that passes only visible light in the chlorophyll absorption band of the reflected light from melon a is disposed diagonally above melon a. The filter 12 has a first filter that passes visible light having a wavelength of about 0.54 μm and a filter that has a wavelength of 0.67 μm.
The configuration is such that one of a second filter and a second filter that passes visible light near μm can be selected (see FIG. 2).

Behind the filter 12, the filter 1
An image input device (camera) 13 composed of an image sensor (sensor) responsive to light passing through 2 is arranged. The image input device 13 is connected to an input side of the image processing computer 4. The image processing computer 4 processes the input image from the image input device 3 as described later. An output side of the image processing computer 4 is connected to an image output device 5 such as a display device or a printer.

Next, a first example of image processing according to the embodiment of the present invention will be described with reference to the flowchart of FIG. In this case, the melon a to be measured is ALS melon. In addition, it is assumed that the filter 12 uses only the second filter that passes visible light having a wavelength of about 0.67 μm.

Now, the visible light emitted from the light source 11 is applied to the melon a, and only the visible light in the chlorophyll absorption band (around 0.67 μm) of the reflected light from the melon a passes through the filter 12. Therefore, an image obtained by photographing with the image input device 13 is formed from chlorophyll information.

Then, the image photographed by the image input device 3 is inputted (S21), and a threshold for extracting the stripes of the skin of the melon a is set (S22). Next, stripes are extracted by binarization (S23), and the extracted stripes are evaluated according to a predetermined procedure (S24).

Next, a second example of image processing according to the embodiment of the present invention will be described with reference to the flowchart of FIG. In this case, the melon a to be measured is ALS melon. The filter 12 selectively uses a first filter that passes visible light having a wavelength of about 0.54 μm and a second filter that passes visible light having a wavelength of about 0.67 μm.

First, the filter 12 has a wavelength of 0.67 μm.
selecting a second filter that passes visible light near m,
At that time, the image captured by the image input device 3 is input and stored (S31). Next, the filter 12 selects a first filter that allows visible light having a wavelength of about 0.54 μm to pass, and at that time, the image captured by the image input device 3 is input and stored (S32).

Further, when the luminance difference between the first image input in step S31 and the second image input in step S32 is obtained (S33), the value of the stripe portion of the skin becomes a positive value, and Is a negative value (see FIG. 2).
Then, after binarizing based on the obtained luminance difference and extracting the stripes and the ground of the epidermis (S34), the extracted stripes are evaluated according to a predetermined procedure (S35).

As described above, in the embodiment of the second aspect of the present invention, the extraction processing of the ground of the melon a and the stripes of the epidermis are performed from the information of the chlorophyll distribution in the epidermis of the melon a. The extraction can be performed with high accuracy.

[0031]

As described above, according to the first aspect of the present invention, the moisture distribution of the epidermis of the test object is detected using near infrared rays in the moisture absorption band, and the surface of the epidermis is detected based on the detection result. And stripes are distinguished from each other, so that the identification accuracy is improved as compared with the related art, and this can contribute to the improvement of the appearance quality test of the inspection object.

According to the second aspect of the present invention, the chlorophyll distribution in the epidermis of the test object is detected using visible light in the chlorophyll absorption band, and the ground and stripes of the epidermis are identified based on the detection result. As a result, the discrimination accuracy is improved as compared with the related art, thereby contributing to the improvement of the appearance quality test of the inspected object.

[Brief description of the drawings]

FIG. 1 is a diagram showing an example of reflection spectral characteristics (spectrum) of near-infrared rays of each part of Aalsmeron.

FIG. 2 is a diagram illustrating an example of reflection spectral characteristics (spectrum) of visible light with respect to each part of the Aalsmeron.

FIG. 3 is a diagram showing an overall configuration of an embodiment of the invention described in claim 1;

FIG. 4 is a flowchart illustrating an example of image processing according to the embodiment.

FIG. 5 is a graph showing an example of a luminance value of an Aalsmeron in a vertical direction.

FIG. 6 is a graph showing an example of a luminance value of the Earls Melon in the horizontal direction.

FIG. 7 is an example of a flowchart for evaluating the crane of a melon with a crane.

FIG. 8 is a diagram illustrating an example of an extracted crane image.

FIG. 9 is a diagram illustrating handling of the image.

FIG. 10 is a diagram showing an entire configuration of an embodiment of the invention described in claim 2;

FIG. 11 is a flowchart illustrating an example of image processing according to the embodiment.

FIG. 12 is a flowchart illustrating another example of the image processing of the embodiment.

FIG. 13 is a plan view illustrating a conventional device.

[Explanation of symbols]

 Reference Signs List 1, 11 light source 2, 12 filter 3, 13 image input device 4, image processing computer 5, image output device

Claims (2)

[Claims] 1. A light source for irradiating near-infrared light toward the skin of a test object, detection means for detecting near-infrared light in a moisture absorption band from reflected light from the test object, and detection of the detection means An appearance evaluation device comprising: identification means for identifying a ground surface of the test object and a stripe portion other than the ground based on the value. 2. A light source for irradiating visible light toward an epidermis of a test object, detection means for detecting visible light in a chlorophyll absorption band from reflected light from the test object, and detection of the detection means An appearance evaluation device comprising: identification means for identifying a ground surface of the test object and a stripe portion other than the ground based on the value.