JPS635157B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention particularly relates to a method for non-contactly measuring the thickness of shiitake mushrooms and automatically sorting them by brand according to predetermined sorting criteria.

In general, in order to sell shiitake mushrooms profitably on the market, whether dried or fresh, it is important to standardize quality standards and select them accordingly.

Conventionally, shiitake mushrooms have been sorted by size using sieves with meshes of various sizes. However, thin meat may be damaged or the outer periphery may be chipped when passed through a sieve, and the thickness and shape cannot be checked with just a sieve, so in the end the meat must be relied on manually. . As described above, manual sorting not only requires high labor costs, but also has many problems such as limited processing capacity and variations in sorting accuracy due to individual differences, and improvements have been desired.

The present invention was made in response to the above-mentioned conventional demands, and its purpose is to indirectly measure the wall thickness of individual shiitake mushrooms without directly touching them, and to meet predetermined selection criteria. To provide an automatic method for automatically sorting shiitake mushrooms, which enables highly efficient and highly accurate sorting by automatically sorting mushrooms.

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

FIG. 1 is a schematic explanatory block diagram showing an embodiment of an automatic sorting device for carrying out the method of the present invention. In the figure, 1 is a measuring conveyor, on which shiitake mushrooms M to be sorted are placed and placed. It is designed to be transported in the direction of the arrow at a speed of .

A camera 2 as an imaging device is installed above the upstream end of the conveyor 1, and is configured to take images of the back surfaces of the shiitake mushrooms M being transported by the conveyor 1. As the image sensor of the camera 2, for example, a CCD or MOS one-dimensional solid-state image sensor is used, and a binary one-dimensional image signal is sequentially captured by this, and a two-dimensional image is constructed by the slice memory unit 3. do.

In addition, an input means that moves a light source and fixes one light receiver or a two-dimensional imaging device may be used.

The number of pixels of the two-dimensional image is, for example, 256 or
The image is 512 bits, and the image shows the folded part m1 and folded part _m2 that appear on the underside of shiitake mushroom M, as shown in Figure _2.
It is composed of two strokes. The scanning start of the camera 2 is performed by an output frequency-divided pulse from a rotary encoder 4 attached to the conveyor 1 and used to detect the amount of movement thereof. In addition, the field of view of camera 2 is the maximum diameter of shiitake mushroom M and conveyor 1.
It is necessary to take into account the variation in the position that can be aligned on the top, but for example, if the width is 150mm, the maximum
It is now possible to sufficiently image even shiitake mushrooms of the same diameter. In order to prevent the camera 2 from capturing two or more mushrooms M at the same time, the shiitake mushrooms M are conveyed one by one onto the conveyor 1. and camera 2
The field of view may be expanded to image a plurality of shiitake mushrooms M at the same time. Furthermore, the camera 2 was scanning in a direction across the conveyor 1, and started capturing image signals the moment it captured the leading end of the shiitake mushroom being transferred, and the image of the shiitake mushroom was cut off after passing the rear end. I am trying to complete it on time.

5 is an arithmetic processing unit that measures the sum of rolled lengths, total circumference length, wrapped area, total area, major axis, minor axis, etc., which will be described later, from the image of the back side of the shiitake mushroom M constructed by the slice memory unit 3. , calculate these measured values, or detect chips and patterns in the outer shape and perform necessary correction processing, and then compare these calculated values with the classification standard setting values to judge the size and sort. It is designed to output a code signal.

Reference numeral 6 denotes a selection standard setting unit for setting the above-mentioned classification standard setting values, and the standard setting values are inputted in advance using a keyboard, a digital switch, or the like.

Reference numeral 7 denotes a pinboard type shift control unit, which is a device for operating the sorting drive unit 8 in accordance with the sorting code signal from the arithmetic processing unit 5.

9 is a video monitor, and 10 is a counting processing device. Reference numeral 11 denotes a lighting device, which uses, for example, a tungsten filament frost bulb.

In the apparatus of the above embodiment, the conveyor 1, camera 2, etc. can be installed in an appropriate number of sets as shown by the imaginary lines in FIG. 1, depending on the processing capacity of the arithmetic unit A.

Next, a method for sorting dried shiitake mushrooms using the apparatus of the above embodiment will be explained.

First, before sending it to the conveyor 1 of this sorting device, for example, large and medium leaves (diameter of 55 mm or more) are sieved.
Middle leaflet (diameter 55-40mm) and leaflet (diameter 40-30mm)
It is preferable to divide the material into parts such as the above and supply them onto the conveyor 1 according to each size, but this is not necessarily necessary.

Next, the camera 2 images the back side of the shiitake mushroom M moving at a predetermined speed on the conveyor 1 as a binary one-dimensional image signal, and the slice memory unit 3 configures this as a two-dimensional image. Next, based on the two-dimensional image of the shiitake mushroom M, the next arithmetic processing unit 5 calculates the rolled circumference ratio, rolled area ratio, minor axis ratio, etc. The wrapped circumference ratio is expressed as the sum of wrapped lengths/total circumference length, the wrapped area ratio is expressed as wrapped area/total area, and the short/major axis ratio is expressed as short axis/long axis. First, from the above image, measure the sum of the rolled lengths, the total circumference length, the wrapped area, the total area, the short axis, and the long axis.

The above-mentioned wrap length sum is the sum of pixels belonging to the boundary between the binarized wrap portion m ₁ and fold portion m ₂ ,
The total circumferential length is calculated as the sum of pixels belonging to the outer circumferential edge of the wrapped portion _m1 . Also, the rolling area is the rolling part
This is the sum of all pixels belonging to m ₁ , and the total area is calculated as the sum of all pixels belonging to both the folded portion m ₁ and the folded portion m ₂ . Furthermore, as shown in Figure 3, the major axis is the two points with the maximum length among the four points Z ₁ , Z ₂ , Z ₃ , and Z ₄ inscribed in the rectangle surrounding the outer shape of the shiitake mushroom M (in this figure, Z ₂ , Z ₄ ), check its periapsis, and choose the longest one as the major axis. Then, the length between two points on the contour line closest to the perpendicular bisector of the major axis is defined as the minor axis.

In addition, if the outer shape of the shiitake mushroom M is missing or the stalk protrudes outward, as shown in Figure 4, connect the points XX or Y-Y where the outline of the outer shape changes rapidly. , and correct the above measured values in advance.

Based on each measurement value calculated as described above, first, it is determined whether or not it is a small piece such as scraps based on the total area, and if it is a small piece, this is ignored and the sorting code signal is not output. I'll keep it. Further, the rolled-up area ratio is calculated from the rolled-up area and the total area to identify the front and back sides of the shiitake mushroom M, and if the rolled-up area ratio is abnormally large, it is regarded as the front side and a return code signal is output, Either the shiitake mushroom M is removed midway and the process is repeated until the back side is detected, or the mushroom M is turned over using an appropriate reversing means and imaged again. In addition, the above-mentioned sorting of small pieces and return operation of shiitake mushrooms with the surface side facing up may reduce the efficiency of the sorting process described in detail below, so this device does not mix small pieces and returns all the mushrooms. An operation process may be performed in advance such that only shiitake mushrooms with their back sides facing upward are flowed onto the conveyor 1.

As described above, from the binarized image of the shiitake mushroom M whose back side is facing upward, not the small pieces, first, the rolled circumference ratio, rolled area ratio, and short/long axis ratio are calculated. The area ratio is calculated by hardware processing in the arithmetic processing unit 5, while the short/major axis ratio is calculated by software processing. Subsequently, the calculated values of the winding circumference ratio, winding area ratio, and minor axis ratio are compared with the classification standard setting values set in advance by the sorting standard setting unit 6. and classify them according to the sorting process shown in FIG. That is, first, the calculated winding circumference ratio is compared with the classification standard set value and classified into three categories: large, medium, and small. As for the wrapped circumference ratio, the smaller the value, the larger the wrapped portion and the thicker the wall. Conversely, the larger the value, the thinner the wall.
As shown in the diagram, we will roughly categorize them into thick, medium, and value.

Next, the calculated winding area ratio is compared with a set reference value, and the water is further divided into small and small categories, which are classified into six categories in total. As for this wrapped area ratio, the larger the value, the larger the wrapped portion and the thicker the wall. Conversely, the smaller the value, the thinner the wall. (average), medium meat (top), medium meat (average),
Classify them into virtue and non-value.

Furthermore, the calculated short/long axis ratio is compared with a set reference value and divided into large and small (large, medium and small at the bottom). As for this ratio of minor to major axis, the larger the value, the more rounded it is, and conversely, the smaller it is, the more elongated it becomes. In this way, the final results are thick (upper), thick (average), medium (upper), medium (average), economical (upper), economical (average), economical (thick), and exceptional. Classify into 8 brands and output the classification code signal.

In addition, when measuring the chipping of the external shape, the degree of chipping (chip area/total area) is also calculated, and if the degree is large, it is preferable to perform a processing operation to move the grade to a lower rank.

The sorting code signal outputted from the arithmetic processing unit 5 is transmitted to the next shift control unit 7, which finally operates the sorting drive unit 8 for sorting.

In the above example, the final classification is into eight categories, but the method of the present invention is not limited to this.
The above-mentioned classification standard setting values can be appropriately set according to the brand selection criteria.

Further, the conveyor is not limited to a linear conveyor as described above, but may be a conveyor such as a turntable, and in short, any conveyor may be used as long as it moves the shiitake mushrooms at a predetermined speed across the field of view of the camera. Therefore, the camera may be moved in the opposite direction.

Furthermore, there is also a method of binarizing an image that uses color signals. For example, the color of the conveyor 1 and the color of the shiitake mushroom M are combined to obtain a signal with the highest contrast ratio. Also, although the image processing is somewhat complicated, the multivalued signal is extracted and a more accurate processing input signal is created.

Furthermore, although the above embodiments have been explained using dried shiitake mushrooms, fresh shiitake mushrooms may also be used.

As described above, the present invention composes an image of a shiitake mushroom without directly touching it using an imaging device, calculates the wrapped circumference ratio and wrapped area ratio of the wrapped portion of the shiitake mushroom from the image, and calculates the wall thickness indirectly. The method uses arithmetic processing to calculate the calculated value necessary for classification, compares it with a predetermined wall thickness classification standard setting value, and operates the sorting drive unit in accordance with the sorting code signal. As a result, there is no risk of damaging the shiitake mushrooms, and it is possible to classify the mushrooms into brands based on their wall thickness very quickly and with high precision.

[Brief explanation of the drawing]

FIG. 1 is a schematic explanatory block diagram showing an embodiment of an automatic sorting device that implements the method of the present invention, and FIG.
An explanatory diagram of a digitized shiitake mushroom image, Fig. 3 is an explanatory diagram of the method for measuring the major axis and minor axis, Fig. 4 is an explanatory diagram of the correction process for irregular images due to defects and stalks, and Fig. 5 FIG. 2 is a diagram illustrating a sorting process. 1... Measuring conveyor, 2... Camera, 3... Slice memory unit, 4... Rotary encoder, 5... Arithmetic processing unit, 6... Sorting standard setting unit, 7... Pin board type shift control unit, 8... Sorting drive unit, 9 ...Video monitor, 10...Counting processing device, 11...Lighting device, A
...Arithmetic device, A...Shiitake mushroom, _m1 ...rolling part, _m2 ...fold part, X, Y, _Z1 to _Z4 ...point.

Claims (1)

[Claims] 1 The back side of the shiitake mushroom M is imaged by the imaging device 2,
A two-dimensional image of the folded part m ₁ and the folded part m ₂ is obtained, and based on this two-dimensional image, the sum of pixels belonging to the boundary between the folded part m ₁ and the folded part m ₂ is defined as the sum of folded lengths, The total circumference is the sum of the pixels belonging to the outer periphery of the wrapped portion m ₁ , and the wrapped area is the sum of all pixels belonging to the wrapped portion m ₁ , which belong to both the rolled portion m ₁ and the folded portion m _2. Each measurement value is calculated using the sum of all pixels as the total area, and if the outer shape of the shiitake mushroom M is missing or the stalk protrudes outward, the outline of the above two-dimensional image changes rapidly2. The above measurement value is corrected in advance by connecting and smoothing the two points, and the ratio of the above-mentioned sum of the winding lengths to the total circumference length is set as the winding circumference length ratio, and the ratio of the winding area to the total area is the winding area. These calculated values are compared with the wall thickness classification standard set value, the magnitude relationship between them is determined, a wall thickness sorting code signal is output, and the sorting drive unit is activated by the sorting code signal. An automatic method for sorting shiitake mushrooms, characterized in that the shiitake mushrooms M are sorted according to their wall thickness.