JPS6261509A - Adjacent seedling detector for rice planter - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention provides a method for detecting the planting state of seedlings planted by a rice transplanter, that is, the planting state, or the adjacent line segments connecting the rows of seedlings. This invention relates to a seedling detection device.

[Conventional technology]

When planting seedlings using a rice transplanter, the seedlings are planted at predetermined intervals in order to facilitate later management and cultivation.
In order to plant the seedlings in a straight line, the already planted row of seedlings adjacent to the aircraft is used as a travel reference line, and the aircraft is steered so that it travels along this travel reference line. Then, an indicator for using the row of already planted seedlings on the outside of the fuselage structure as the aircraft travel target line is provided in a state that protrudes toward the outside of the fuselage structure, and the positional relationship between this indicator and the row of already planted seedlings is visually checked. By controlling the aircraft while checking the planting conditions, the relationship between the seedlings and the position of the aircraft was maintained properly.

(For example, see Japanese Utility Model Application Publication No. 57-82109).

[Problem that the invention seeks to solve]

However, in the above-mentioned conventional means, since the indicator that moves on the row of already planted seedlings on the outside of the aircraft structure as the aircraft moves is visually confirmed, errors due to the influence of parallax and local planting may occur. This has the disadvantage that misidentification of the position due to unevenness of the seedling, missing plants, etc. is likely to occur, resulting in inaccurate results.

Therefore, a method has been considered for planting adjacent to the aircraft by contacting the seedlings and installing a contact type sensor that detects the positional relationship between the seedlings and the aircraft. Since the strength of the contact sensor is weak, it is necessary to make the contact piece of the contact sensor very weak in order to avoid pushing down the seedlings during planting using the contact sensor. If the contact piece strength of the contact sensor is made very weak, the sensor's sensitivity to detecting seedlings will be high, and even the slightest vibration will act as if it were in contact with the seedlings.
This has the disadvantage that it is prone to malfunction.

On the other hand, non-contact planting is a means of detecting the positional relationship between the seedlings and the aircraft. Although methods for detecting the position of seedlings have been considered, these contact-type,
With any of the non-contact methods, the position of the seedling is detected for planting, and the position of the seedling is detected at a single point on the surface for planting in the field. It has the disadvantage that it is easily influenced by the strain and the detection surface position is inaccurate.

The present invention has been made in view of the above-mentioned circumstances, and its purpose is to enable the state of seedlings to be detected in a non-contact manner during planting, and as a line segment connecting multiple seedlings during planting. There is a particular thing.

[Means for solving problems]

The characteristic configuration of the adjacent seedling detection device for a rice transplanter according to the present invention is as follows:
For planting a plurality of seedlings in a predetermined range including seedlings, an imaging means for taking an image of a surface, a color difference calculation means for calculating a color difference between a green component and a blue component of the image information captured by the imaging means, and a color difference calculation means based on the color difference calculation means are used. By binarizing the color difference image information based on a set threshold value, the area corresponding to the seedling is extracted for planting2.
The present invention is provided with a value converting means and an adjacent seedling row detecting means for linearly approximating the line segment connecting the extraction area by the binarizing means, and their functions and effects are as follows.

[For production]

As shown in the table below, in general, the color of seedlings during planting is greenish, so when planting in a field, if we focus on the green component (G) of the image information of the seedling state, the image part is It looks bright because it has a lot of green components, but in areas where the mud surface of the field is exposed, it looks dark because there is little green component.

However, since almost all natural light is reflected on the water surface, the reflected light contains all color components, making the water surface appear brighter. However, in the above-mentioned planting, the seedlings are greenish and the muddy surface is brownish or grayish, so the blue component contained in the color components is small. Therefore, if we pay attention to the blue component (B) of the imaged information, we can see that the reflected light from the water surface contains all the color components, so the water surface appears bright (seems to be bright), while the seedlings and mud surface of the planted plants appear dark. becomes.

Therefore, the image information of the green component (G) and the blue component (B
) of the color difference image information (G
-B) is for planting only seedlings.

The present invention utilizes the above-mentioned characteristics, and when planting a plurality of seedlings, images are taken of a surface in a predetermined range including seedlings, and the color difference between the green component and the blue component of the imaged information is calculated. For planting, the color difference image information corresponding only to the color of the seedlings is extracted, and the color difference image information is binarized based on a set threshold value to remove noise components contained in the image information. Each area of the binarized image information corresponds to planting only seedlings. Then, by connecting each area of this binarized image information using a straight line approximation, even if there are locally curved plantings, overturned seedlings, or missing plants, the entire area will not be affected. For systematic planting, line segments can be obtained that correspond to the condition of the seedlings.

[Effect of the invention] In other words, planting multiple seedlings, including seedlings, can be done by binarizing the image information obtained by calculating the difference between the green component and the blue component of the image information obtained by imaging the state of the surface. Since image information corresponding only to the seedlings can be obtained, it is possible to clearly distinguish between the background of the water surface, mud surface, etc. and the planted seedlings. Then, the extracted planting connects the areas corresponding to the seedlings by linear approximation, so the planting is not affected by local unevenness or missing plants of the seedlings, and the overall planting is information that corresponds to the state. It can be obtained precisely.

(Example〕 Embodiments of the present invention will be described below based on the drawings.

As shown in Figures 6 and 7, the device (2) for planting seedlings is installed at the rear of the aircraft (V), which is configured to be able to steer both the front wheels (IF) and rear wheels (IR). In addition to being movable up and down, a color video camera (3) is provided as an imaging means for imaging the state of already planted seedlings adjacent to the body (v), and the image information captured by this camera (3) is transferred to an adjacent seedling detection device to be described later. (A), and to plant multiple seedlings (T), the positional relationship of the aircraft (V) with respect to the line segment connecting the seedlings (T), that is, the traveling reference line (L), is detected, and this detected traveling reference '1IA (L ) Based on the detection results of the travel control device (
H), the vehicle body (V) is configured to perform a steering operation to automatically travel along the travel reference line (L).

The camera (3) is installed at the tip of the support frame (4) that protrudes toward the outer side of the fuselage (V) so as to image the row of already planted seedlings adjacent to the outer side of the fuselage structure from diagonally above. When the aircraft (V) is along the planted seedling row, that is, the running reference line (L) at a set distance, this running reference line (L) is above and below the imaging field of view of the camera (3). It is arranged so that it is located at the center of the direction (indicated by (Lo) in the figure).

As shown in Figure 5, the driving force from the engine (E) is
The front wheels (IF) and the rear wheels (IF) are connected to each other via the transmission (4).
IR), and the motor (5) as an actuator for shifting operation is transmitted so that the shifting position detected by the shifting position detecting potentimeter (R3) becomes a predetermined position.
The vehicle is configured to automatically travel at a predetermined speed by driving the vehicle.

Further, the front wheels (IF) and the rear wheels (IR) are configured to be power-steering operated by hydraulic cylinders (6F) and (6R), respectively, and a steering angle detection potentiometer is connected to the steering operation of the wheels. (R,), (R1) so that the detected steering angle matches the target steering angle.
1'), solenoid valve (7F) that operates (6R)
, (7R).

Then, the aircraft (v) with respect to the detected traveling reference line (L)
When correcting the tilt (α) in the longitudinal direction, adjust the aircraft (V)
Front wheel (IF) and rear wheel (IR) to change direction
The aircraft (V) is rotated relative to the traveling reference line (L) by performing turning steering in a relatively opposite direction.
When correcting the deviation (β) in the width direction, the front wheels (IF
) and the rear wheels (IR) in the same direction.
) is efficiently controlled to follow the travel reference line (L).

The configuration and operation of the adjacent seedling detection device (A) will be described below.

As shown in FIG. 1, the adjacent seedling detection device (A) includes NTSC images captured by the color video camera (3) and output as image information (see FIG. 4 (a)).
format color video signal (So), vertical synchronization signal (V
D) NT that separates each synchronization signal, horizontal synchronization signal (HD), and each color signal, green signal (G), blue signal (B)
SC decoder (8), the green signal (G) and the blue signal (
A color difference calculation unit (20) that calculates the difference (G-8) of B) in the state of an analog signal, and a color difference signal (G-8) as color difference image information (Sl) output from this color difference calculation unit (20).
A binarization processing unit (30) binarizes the data based on a set threshold value (Cref), and the planting extracted by this binarization processing unit (30) corresponds only to the color (C) of the seedling (T). The binarized image information (F+) (see Figure 0 (b)) is divided into 32 x 3
Image memory (M) that stores image information consisting of 2 pixels/1 screen, 2 pixels stored in this image memory (M)
A line segment (L) connecting each area (Fa) of the digitized image information (F1)
) (see Figure 4 (c)) is subjected to Hough transform and linearly approximated, and the slope (a) and deviation (b) of the linearly approximated line segment (L) with respect to the center of the imaging field of view (Lo) are calculated. A Hough transform unit (40) as a seedling row detection means, a control unit (50) that controls the operation of each unit, a monitor (1O) that displays captured image information (S1) and calculation results of each unit, and the vertical synchronization unit. A CRT controller (11) is provided which receives a signal (VD) and a horizontal synchronization signal (HD) and generates various control signals for synchronizing the operations of the monitor (10) and other parts. In addition, in FIG. 1, (H) is a travel control device.

The configuration of each part will be explained in detail below.

As shown in FIG. 2, the color difference calculation unit (20) has its operation controlled by a timing controller (21) so that it operates in synchronization with the horizontal synchronization signal (HD) and the vertical synchronization signal (VD). It is intended to be done. and,
The difference (G-B) between the green component color signal (G) and the blue component color signal (B) output from the NTCS decoder (8)
) is calculated by a differential amplifier (22), the output signal is amplified to a predetermined level by a video amplifier (23) whose amplification degree is set by a gain setting resistor (Rg), and a comparator ( 30) by resistance (Rf)
compared with the setting threshold (Cref) set by
The captured image information (So) shown in FIG. 4(A) is converted into TTL level binary image information (F1) shown in FIG. 4(B). In other words, the planting area (Fa ) is obtained in real time as the image is captured. In addition, the total threshold value (Cre
f) is set in advance based on the color of the recognition target bacterium (T), the brightness of the imaging state, etc.

The Hough transform unit (40) converts the coordinates (x, y) for each coordinate (x, y) of each area (Fa) of the binarized image information (F1) stored in the image memory (M). The line segment (L) passing through is defined as the Hough value (ρ) and its angle (θ) expressed in the polar coordinate system based on the following formula (i), and the same Hough value (ρ) is taken. The frequency is counted as a two-dimensional histogram, the line segment (L) is specified from the maximum Hough value (ρ) and its angle (θ), and the Hough value corresponding to the specified line segment (L) is calculated. After receiving information on the value (ρ) and its angle (θ), an image center line (Lo ) to find the slope (a) and deviation (b).

ρxx sin θ + y cos θ −− (i ) y tomi a −
x+b...(ii) Here, a-
f(θ) and b−f(ρ).

Next, the overall operation of the adjacent seedling detection device (A) having the above configuration will be explained based on the flowchart shown in FIG.

That is, the image is captured by the camera (3), and the NTSC
The video signal for one screen converted into a green component color signal (G) and a blue component color signal (B) by the decoder (8) is input to the color difference calculation section (20), and the color difference is calculated in real time. (G-8) is calculated (step #1).

The color difference (GB) signal is sent to the binarization processing unit (30
) and stored in the image memory (M) (
Step 12).

Binarized image information (F) stored in the image memory (M)
l) is transformed by the Hough transform unit (40) into each region (
A two-dimensional histogram of the Hough value (ρ) and its angle (θ) of the line segment (L) passing through Fa) is obtained (Steflo 3
), the line segment with the maximum value of the two-dimensional histogram is identified as the running reference line (L) connecting adjacent seedling rows, the inclination (a) and deviation (b) on the image are detected, and the control unit ( 50
) is sent to the traveling control device (H) as a traveling control parameter (step 14).

The travel control device (H) determines the slope (a) of the travel reference line (L) on the image as a travel control parameter.
and deviation (b), calculate the inclination (α) and deviation (β) of the aircraft (V) in the width direction with respect to the running reference line (L), and based on the calculation results, move the aircraft (V). A steering operation is performed (step f15).

In the above embodiment, the reference position for the inclination (a) and deviation (b) of the traveling reference line (L) connecting adjacent seedling rows was set as the line (L1) that vertically traverses the vertical center of the imaging field of view. Any position on the image, such as the left, right, top, and bottom ends or the center point, may be used as the reference.

Further, in configuring the Hough transform unit (40), in addition to calculating the above-mentioned equations (i) and (ii) by arithmetic processing,
The Hough value (ρ) corresponding to each angle obtained by dividing the polar coordinate angle (θ) into a predetermined range is calculated in advance, and the results are stored in a table for each coordinate (x, y), so-called table look. The Hough transform process may be performed using an amplifier method. In this case, there is no need to perform complicated arithmetic processing involving polar coordinate transformation, and the Hough transform process can be performed at high speed.

[Brief explanation of drawings]

The drawings show an embodiment of the adjacent seedling detection device for a rice transplanter according to the present invention, FIG. 1 is a block diagram showing the configuration of the adjacent seedling detection device, FIG. 2 is a block diagram showing the configuration of the color difference calculation section, Figure 3 is a flowchart showing the operation of the adjacent seedling detection device, Figure 4 (A) is an explanatory diagram of captured image information, Figure 4 (B) is an explanatory diagram of binarized image information, and Figure 4 (C) is a straight line. Illustration of approximation,
FIG. 5 is a block diagram showing a schematic configuration of the travel control system of the rice transplanter, FIG. 6 is an overall plan view of the rice transplanter, and FIG. 7 is an overall side view of the rice transplanter. (3)... Imaging means, (20)... Color difference calculation means, (30)... Binarization means, (40
)... Adjacent seedling row detection means, (T)...
Planting is seedlings, (So)...Image image information, (G)
...Green component, (B) ...Blue component,
(GB)...color difference, (Fa)...area, (L)...line segment connecting extraction areas.

Claims (1)

[Claims] An imaging means (3) for imaging a predetermined range of seedling planting surfaces including a plurality of planted seedlings (T), a green component (G) and a blue component (B) of image information (S_0) captured by the imaging means (3);
) color difference calculation means (20) for calculating the color difference (GB) of
Color difference image information (S_1) obtained by this color difference calculation means (20)
) based on a set threshold value to extract an area (Fa) corresponding to the color of the planted seedling (T);
An adjacent seedling detection device for a rice transplanter, comprising adjacent seedling row detection means (40) for linearly approximating a line segment (L) connecting Fa).