JP3570765B2 - Seeding evaluation device - Google Patents

Description

[0001]
[Industrial applications]
The present invention is an imaging unit that images a seeding surface on which a plurality of crop seeds are sown, and a seed region extraction unit that extracts a seed region corresponding to the seed based on imaging information of the imaging unit. The present invention relates to a sowing evaluation apparatus provided with an evaluation unit for evaluating a seed sowing state of the seed on the seed sowing surface based on information of the seed region extracting unit.
[0002]
[Prior art]
Conventionally, the seeding evaluation device extracts a seed region by binarization processing on a seed sowing surface, calculates the number of the seed regions, and evaluates whether the seed sowing distribution is in an appropriate state (for example, And JP-A-63-309107.
[0003]
[Problems to be solved by the invention]
However, in the above-described conventional technique, since the image region extracted corresponding to the seed by the binarization processing on the seed seeding surface is used as the seed region as it is, a plurality of seeds are close to each other and extracted as one image region. In such a case, the number of seeds that is originally plural is erroneously calculated as one, and there is a problem that a proper seeding state cannot be evaluated.
[0004]
The present invention has been made in view of the above circumstances, and an object of the present invention is to extract a plurality of seeds close to each other on a seeding surface as one image area in order to solve the above-described disadvantages of the related art. Even in this case, the number of seeds that is originally plural is calculated as the original number, and the proper seeding state is evaluated.
[0005]
[Means for Solving the Problems]
A first characteristic configuration of the seeding evaluation device according to the present invention is that the seed region extracting unit extracts an image region after subjecting the image region corresponding to the seed to a predetermined amount to be reduced as the seed region, but it has line evaluation of the state sown the seed based on the number information of the seed region after the reduced-treated,
The imaging means (S) is configured to move in a two-dimensional direction in a horizontal plane, and sequentially capture images of each range (1a) obtained by dividing the seeding surface in a vertical direction and a horizontal direction. .
[0006]
In a second characteristic configuration, in the first characteristic configuration, the evaluation unit divides the entire seed sowing surface into a plurality of small regions, and obtains the seed density information for each of the small regions. The present invention is configured to evaluate the seed sowing state based on the evaluation.
[0007]
In a third feature configuration, in the second feature configuration, the size of the small region is made to correspond to a stocking range when the crop grown on the seeding surface is stocked and planted in a field. On the point.
[0008]
[Action]
According to the first characteristic configuration of the sowing evaluation device of the present invention, an image area corresponding to a seed is obtained in a captured image of a seed sowing surface on which a plurality of crop seeds are sown, and the image area is further reduced by a predetermined amount. Image regions separated by body processing and corresponding to each seed are extracted as seed regions. Then, based on the number information of the seed regions corresponding to the respective seeds, the state of the seed sowing on the seed sowing surface is evaluated based on, for example, whether or not the number of the seed regions is within an appropriate number range.
[0009]
According to the second feature configuration, in the first feature configuration, the entire seed sowing surface is divided into a plurality of small regions, and each of the small regions is separated and extracted so as to correspond to each of the seeds. Based on the number of seed regions thus obtained, that is, the density information, the seed sowing state on the seed sowing surface is evaluated based on, for example, the positions of high and low seed densities and variations in the density.
[0010]
According to the third feature configuration, in the second feature configuration, the size of the small region corresponds to a range of stocking when crops grown on a seed sowing surface are stocked and planted in a field. For example, it is possible to predict the growth density of the crop when the plant is divided and planted in the field from the variation in the density between the high density portion and the low density portion of the seed in each small region.
[0011]
【The invention's effect】
Therefore, according to the first characteristic configuration of the seeding evaluation device of the present invention, even when a plurality of seeds are close to each other on the seeding surface and are extracted as one image region, the one image region is extracted. Since the seed regions are separated and extracted as seed regions corresponding to the respective seeds, it is possible to evaluate a proper seeding state based on the number information of the seed regions corresponding to the respective seeds.
[0012]
Further, according to the second characteristic configuration, the seeding state can be finely evaluated not for the entire seeding surface but for each part thereof, and thus suitable means of the first characteristic configuration can be obtained. You.
[0013]
According to the third feature structure, and Kabuwake crops grown on the seed播面, suitable state can be evaluated whether the like can be used to instill in the field, have been, the first feature structure Ru suitable means to obtain the.
[0014]
【Example】
Hereinafter, a case where an embodiment of the present invention is applied to a sowing facility for growing rice seedlings as crops will be described with reference to the drawings.
[0015]
As shown in FIG. 1, a color as imaging means for imaging the upper surface of the tray, that is, the seed sowing surface, above a rectangular seedling raising tray 1 on which a plurality of rice seeds T have been sown in a preceding step (not shown). A CCD camera S is installed so as to be movable in a two-dimensional direction in a horizontal plane.
[0016]
The tray 1 is set so as to be at a predetermined position with respect to the CCD camera S, and the tray 1 is set so that the area 1a obtained by dividing the seeding surface of the tray 1 into six in the vertical and horizontal directions corresponds to the imaging field of view of the camera. The distance from the camera S is set, and the CCD camera S is operated to move so as to sequentially capture images in the respective ranges 1a.
[0017]
As shown in FIG. 2, a control device H using a microcomputer is provided, and the control device H includes a micro switch for detecting the imaging information of the CCD camera S and detecting that the tray 1 is set at the predetermined position. And other detection information from the tray set switch SW. On the other hand, the control device H outputs a drive signal for a movement motor M for moving the CCD camera S. The rotational force of the moving motor M is converted into a force for moving the CCD camera S via a mechanism such as a rack and pinion (not shown).
[0018]
Using the control device H, based on the imaging information of the CCD camera S, a seed region extracting unit 100 for extracting a seed region corresponding to the seed T, and based on information of the seed region extracting unit 100, An evaluation unit 101 configured to evaluate a seed sowing state of the seed T on the seed sowing surface is configured. Then, the seed area extracting means 100 extracts an image area obtained by subjecting the image area corresponding to the seed T to a predetermined amount of reduction processing as the seed area. The method is configured to evaluate the seed sowing state based on the number information of the seed regions.
[0019]
The seed region extracting means 100 will be described. As shown in FIG. 3A, the difference in color or brightness of the white seed T with respect to the black soil B for each pixel having a predetermined resolution in the vertical and horizontal directions of the screen. (The pixel value corresponding to the soil B is set to 0 and the pixel value corresponding to the seed T is set to 1), and an image area Ta ′ corresponding to the seed T is extracted. Here, each image region Ta ′ usually corresponds to one seed T, but since a plurality (for example, two) of seeds T may be close to one image region Ta ′, FIG. As shown in FIG. 4, the image region Ta ′ is subjected to a contraction process to extract image regions Ta separated into individual seeds T1 as seed regions. In the figure, the position G of the center of gravity of the various regions Ta is further obtained.
[0020]
Specifically, as shown in FIG. 5, the above-mentioned contraction processing is performed on each pixel g (i, j) (i = 1 to the number of pixels in the horizontal direction, j = 1 to the number of pixels in the vertical direction) of the entire screen. If at least one pixel value of eight pixels surrounding the target pixel g (i, j) is 0, the value of the pixel g (i, j) is set to 0, and all the pixel values of the eight pixels are Only when the value is 1, a process of holding the value of the pixel g (i, j) as it is is performed. One contraction processing is completed as described above. For example, in a condition in which it is not possible to separate each T1 image area Ta by one contraction processing, two or more contraction processings may be performed as necessary. Do.
[0021]
Further, the evaluation means 101 divides the entire seed sowing surface into a plurality of small areas, and evaluates the seed sowing state based on the density information of the seed T for each of the small areas. Then, as shown in FIG. 3 (b), the size of the small region k is determined by dividing the seedlings of rice grown on the seeding surface into strains and planting them in a field (for example, a rice planting machine). 17 mm square corresponding to the planting nail). The density of the seed T is obtained assuming that the various regions Ta belong to the small region k where the center of gravity G is located. Incidentally, the state of the density in which about 3 to 5 seeds T exist in the small area k is an appropriate seed sowing state.
[0022]
Next, the seeding evaluation process performed by the control device H will be described with reference to the flowchart in FIG. First, it is confirmed whether or not the tray 1 is set at a predetermined position based on the information of the tray set switch SW. If the tray 1 is set at a predetermined position, an image in the range 1a of the first seeding surface is captured at a predetermined screen resolution (pixel density), and the captured image is subjected to the above-described binarization processing and contraction processing. Is performed, and various regions Ta are extracted by performing a labeling process, and then the center of gravity G of the various regions Ta is obtained. Then, the number of seeds T, that is, the density, for each of the small regions k is obtained, and whether or not the seeding density is appropriate for each small region k is evaluated.
[0023]
Next, when the seeding evaluation for one tray 1 has not been completed, the camera is moved to the range 1a of the second and subsequent seeding surfaces, and the above processing is repeated. When the seeding evaluation for one tray 1 is completed, the evaluation result is displayed on a display screen or the like and stored, and the process waits for the next tray 1 to be set.
[0024]
(Another embodiment)
In the above embodiment, the case of growing rice seedlings as crops has been described, but the invention can be applied to various crops other than rice. Further, in order to form a seed sowing surface for sowing the seed T of the crop, the tray 1 does not necessarily need to be the tray 1, and a seeding surface having a larger area may be used.
[0025]
In the above embodiment, the image pickup means is constituted by the color CCD camera S. However, the present invention is not limited to this, and another image pickup means such as a monochrome CCD camera may be used.
[0026]
In the above embodiment, the contraction processing is performed based on the pixel values of eight pixels surrounding the target pixel g (i, j). However, the present invention is not limited to this, and the target pixel g (i, j) is not limited to this. Alternatively, the determination may be performed based on the pixel values of four pixels adjacent in the horizontal and vertical directions of the screen, or the pixel values of two pixels adjacent in the vertical or horizontal direction of the screen.
[0027]
In the above embodiment, the evaluation means 101 divides the entire seed sowing surface into a plurality of small areas, and evaluates the seed sowing state based on the density information of the seed T for each of the small areas. However, it is not necessary to divide the seeds into a plurality of small areas. Based on the number information of the seeds T on the entire seeding surface, for example, it is determined whether the number of the seeds T is an appropriate number or not. Is also good.
[0028]
Also, in the case where the entire seed sowing surface is divided into a plurality of small regions, the size of the small region is determined by dividing the crop grown on the seed sowing surface into a stocking range k when planting in a field. It is not necessary to make the size of the small area appropriately set according to the specific use form of the seeding evaluation result.
[0029]
In the claims, reference numerals are provided for convenience of comparison with the drawings, but the present invention is not limited to the configuration of the attached drawings by the entry.
[Brief description of the drawings]
FIG. 1 is a perspective view showing an imaging state of a seed sowing surface by an imaging unit. FIG. 2 is a block diagram of a control configuration of a sowing evaluation apparatus. FIG. FIG. 5 is an explanatory diagram of a somatic process. FIG. 5 is an explanatory diagram of a contraction process. FIG. 6 is a flowchart of a control operation of the seeding evaluation device.
T type S imaging means 100 type area extracting means 101 evaluation means

Claims (3)

Imaging means (S) for imaging a seed sowing surface on which a plurality of crop seeds (T) are sown; and a seed region corresponding to the seed (T) based on imaging information of the imaging means (S). A seed region extracting unit (100) to be extracted; and an evaluating unit (101) for evaluating a seed sowing state of the seed (T) on the seed sowing surface based on information of the seed region extracting unit (100). A sowing evaluation device provided,
The seed region extracting means (100) extracts, as the seed region, an image region after subjecting the image region corresponding to the seed (T) to a predetermined amount of contraction processing;
Said evaluating means (101) is, have rows evaluation of status sowing the seed on the basis of the number information of the seed region after the reduced-treated,
A sowing evaluation device configured so that the imaging means (S) moves in a two-dimensional direction in a horizontal plane and sequentially captures an image of each range (1a) obtained by dividing the seeding surface in a vertical direction and a horizontal direction. .   The evaluation means (101) divides the entire seeding surface into a plurality of small areas, and evaluates the seeding state based on density information of the seed (T) for each of the small areas. The seeding evaluation device according to claim 1, which is configured as follows.   The seeding evaluation device according to claim 2, wherein the size of the small region corresponds to a range in which the crop grown on the seeding surface is divided and planted in a field.

Images