JPH06335662A - Device for evaluating appearance of melon - Google Patents

Abstract

PURPOSE:To attain the fairness in the evaluation of the mesh structure of melons and to lessen the manpower for this evaluation. CONSTITUTION:The image of the melon (a) is inputted S1 and after the meshes (stripes) of the melon are extracted S2 from this input image, plural pieces of the basic figures included in these extracted meshes are set S3. The number of the basic figures included in the extracted meshes is then calculated S4. In succession, the inclination of a straight line is determined S5 on logarithmic graphs by a method of least squares in accordance with the basic figures and the number of the basic figures included in the calculated meshes. The absolute value of the inclination of the straight line indicates fractal dimension and is the index of the intricateness of the structure of the meshes to be evaluated, i.e., the density of the meshes. The class of the melon is, thereupon, decided S7 according to the structure of the meshes by referring S6 to a table determining the relation between the inclination of this straight line and the classes of the melons.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a melon appearance evaluation apparatus for evaluating the structure of a net-type melon such as musk melon, yubari melon and andes melon which has a mesh (stripes) on the surface of the epidermis.

[0002]

2. Description of the Related Art In net melons, the quality of the mesh formed on the surface is regarded as an important quality factor for appearance. Therefore, conventionally, the inspector visually evaluates the mesh tension of the melon (complexity of the mesh structure) and evaluates it.

[0003]

As described above, conventionally, since the inspector must evaluate the tension of the melon mesh, the evaluation is uneven and lacks fairness. Labor saving was desired.

Therefore, an object of the present invention is to achieve fairness in the structure evaluation of the melon mesh and to save labor in the evaluation work.

[0005]

In order to achieve the above object, the present invention has the following constitution.

That is, according to the present invention, an image pickup means for photographing a melon having a mesh, a mesh extracting means for extracting a mesh of the melon from the photographed image, and a basic figure constituting the extracted mesh are included in the mesh. The number of basic figures to calculate the number of basic figures, the fractal dimension calculating means for calculating the fractal dimension of the mesh from the calculated number of basic figures included in the basic figure and the mesh, and the fractal dimension of the melon based on the calculated fractal dimension. And a mesh evaluation means for evaluating the structure of the mesh.

[0007]

In the present invention, the image pickup means photographs the melon having the mesh, and the mesh extracting means extracts the mesh of the melon to be evaluated from the photographed image. The basic figure number calculating means calculates the number of basic figures forming the extracted mesh included in the mesh.

The fractal dimension calculating means calculates the fractal dimension of the mesh from the basic figure and the calculated number of basic figures included in the mesh. The obtained fractal dimension is an index of the complexity of the mesh structure, in other words, the denseness of the mesh.
Therefore, the mesh evaluation means evaluates the structure of the melon mesh based on the obtained fractal dimension.

As described above, in the present invention, based on the knowledge that the structure of the melon mesh has fractal properties, the fractal dimension of the mesh is obtained to quantitatively evaluate the structure of the melon. The fairness of the structural evaluation can be achieved and labor saving of the evaluation can be achieved.

[0010]

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

In FIG. 1, reference numeral 1 denotes an image input device composed of an image pickup device and the like, which is arranged laterally of the melon a in order to photograph the side surface of the melon a. This image input device 1
Is connected to the input side of the image processing apparatus 3 via the camera controller 2. The image processing device 3 includes a CPU, a memory, and the like, and performs a predetermined process on an input image from the image input device 1 in a predetermined procedure as described later. Image processing device 3
The display device 4 is connected as an image output device to the output side of. In addition, the image processing device 3 includes a system controller 5
Connect with.

Next, an example of image processing of the embodiment thus constructed will be described with reference to the flowchart of FIG.

First, the melon a photographed by the image input apparatus 1
Image is input (S1). Next, a melon mesh (stripes) is extracted from the input image (S2), and then a plurality of basic figures forming the mesh are set (S3). This basic figure is, for example, a square of 4, 16 or 64 pixels.

Next, for example, as shown in FIG. 3, the number of basic figures included in the extracted mesh is calculated by the overlap (hatched portion) of the basic figure and the mesh (S4). Now
The basic figures are 4, 16, and 64 pixels, and each of these basic figures (the length r of one side of the basic figure) is on the horizontal axis, and the number N (r) of each of the basic figures included in the mesh is on the horizontal axis. By taking the above, the measurement result as shown in FIG. 4 or 5 is obtained.

The percentage shown in this graph indicates the ratio of the basic figures included in the mesh. For example, in the case of FIG. 3, the ratio is 75%. It should be noted that FIG. 4 shows a case where the mesh of the melon is wholly dense, and FIG. 5 shows a case where the mesh is wholly rough.

Subsequently, based on the result shown in FIG. 4 or 5, the slope of the straight line is obtained on the log-log graph by the least square method (S5). The absolute value of the slope of this straight line indicates the fractal dimension, and the complexity of the mesh structure to be evaluated,
That is, it is an index of the fineness of the mesh. Then, the table that defines the relationship between the inclination of the straight line and the grade of melon is referred to (S6), and the grade of melon is determined according to the mesh structure (S7).

Next, the appearance of the melon is evaluated comprehensively by using a plurality of evaluation factors such as the unevenness of the skin of the melon and the rise of the mesh (stripes) in addition to the mesh structure of the melon obtained as described above. The method will be described below.

In this method, the network structure of the melon is quantified by the fractal dimension as described above, the unevenness of the skin of the melon is quantified by the degree of deformation of the figure obtained by laser irradiation on the surface of the melon, and the rise of stripes is the melon. It is quantified by measuring the displacement of the surface.

Then, based on these digitized values,
Perform a comprehensive evaluation of melon appearance using fuzzy reasoning. That is, as the antecedent part of the fuzzy rule, the mesh structure of the melon (X1), which is the evaluation element, the unevenness of the skin of the melon (X
2) and the swelling of the mesh (X3), and then the consequent part is the comprehensive evaluation value (Y). Then, fuzzy inference is performed using the fuzzy rules as shown below, and a comprehensive evaluation of the appearance of the melon is performed.

An example of fuzzy rules will be shown below.

{IF X1 = ZO, X2 = ZO, X3 =
ZO THEN Y = ZO is excellent if all are the same} {IF X1 = NS X2 = ZO, X3 = ZO THE
N Y = ZO Even if the mesh is slightly weak, it is excellent if everything else is the same} {IF X1 = PS, X2 = ZO, X3 = ZO THE
N Y = PS Even if everything else is the same if the mesh is dense, it is well downgraded} With the above evaluation method, the appearance of the melon was evaluated comprehensively based on multiple evaluation factors. It is possible to quantitatively and stably perform the same comprehensive evaluation as that performed by a person based on visual observation.

[0022]

As described above, according to the present invention, the fractal dimension of the mesh is obtained and the structure of the mesh is quantitatively evaluated based on the knowledge that the structure of the melon mesh has fractal properties. Therefore, the fairness of the structure evaluation of the melon mesh can be achieved, and the labor saving of the evaluation can be achieved.

[Brief description of drawings]

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

FIG. 2 is a flowchart showing an image processing example of the embodiment.

FIG. 3 is a diagram illustrating a relationship between a mesh and a basic figure.

FIG. 4 is a graph showing an example of a relationship between a basic figure and the number of the basic figure included in a mesh.

FIG. 5 is a graph showing another example of the relationship between the basic figure and the number of the basic figure included in the mesh.

[Explanation of symbols]

 a Melon 1 Image input device 3 Image processing device 4 Display device

[Procedure amendment]

[Submission date] August 28, 1992

[Procedure Amendment 1]

[Document name to be amended] Statement

[Correction target item name] 0016

[Correction method] Change

[Correction content]

Subsequently, based on the result shown in FIG . 4 or 5 , the slope b / a (see FIG. 6) of the straight line is obtained on the logarithmic log graph by the method of least squares (S5). Absolute value of the slope of the straight line represents the fractal dimension, the complexity of the structure of the mesh cells for evaluation, i.e. Ri Do indicative of compactness of the mesh, its
The larger the value of, the more complicated the mesh . Then, the table that defines the relationship between the slope of the straight line and the grade of melon is referred to (S6), and the grade of melon is determined according to the mesh structure (S7). By the way, as mentioned above, the melon mesh
When the whole is dense, the data shown by chair 4 is obtained.
When the melon mesh is coarse as shown in Fig. 5,
Such data can be obtained. Therefore, as shown in FIG.
The straight line obtained in step S5 and the preset vertical axis
Find the intersection that intersects with the parallel straight line c, and from that intersection, the horizontal axis
The density of the mesh is evaluated by the value e on the vertical axis that has moved parallel to
it can. The larger the value e, the denser the mesh.

[Procedure Amendment 2]

[Document name to be amended] Statement

[Name of item to be corrected] Figure 6

[Correction method] Added

[Correction content]

FIG. 6 is a diagram for explaining a mesh evaluation method.

[Procedure 3]

[Document name to be corrected] Drawing

[Name of item to be corrected] Figure 6

[Correction method] Added

[Correction content]

[Figure 6]

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 7/18 B

Claims (1)

[Claims] 1. An image pickup means for photographing a melon having a mesh, a mesh extracting means for extracting a mesh of the melon from the photographed image, and a number of basic graphics constituting the extracted mesh included in the mesh. And a fractal dimension calculating means for obtaining the fractal dimension of the mesh from the basic figure and the calculated number of basic figures included in the mesh, and the structure evaluation of the melon mesh based on the obtained fractal dimension. An apparatus for evaluating the appearance of melon, comprising: