JPH02301809A - Picture processor for mobile work machine - Google Patents

Abstract

PURPOSE:To reduce the processing time of the processor by establishing the specific range of one edge line based on converted values of candidate points of the corner of one side edge and specific values of another approximate edge line and detecting the specific value of the edge line. CONSTITUTION:During the course of Hough transformation process, specific values rho1 and theta1 of one approximate edge line A1 are detected based on the candidate points xn and yn of one side having high sharpness. Then, by establishing the specific range of another edge line A2 based on the converted value S of a corner candidate point xc and yc of the one side edge and the specific values rho1 and theta1, the specific values rho2 and theta2 of the edge line A2 are detected. Therefore, specification of an approximate edge line by performing Hough transformation on each edge candidate point can be performed surely at a high speed, resulting in the improvement in throughput.

Description

Detailed Description of the Invention "Industrial Application Field" The present invention relates to an image processing device for a mobile work machine such as a combine harvester, which is equipped with a camera that images a field and an image processing device that processes this image, and which allows the machine to travel unmanned and automatically. Regarding.

``Prior art'' When image information as shown in Fig. 5 is obtained from a camera in the direction of movement, when detecting the boundary edge line between the rice cultivation area and the non-cultivation area, the vertical edge line also becomes the horizontal edge line. In the case of , for example, detection is performed by scanning and searching for a portion of the screen after binarization where the change in shading is large.

"Problem to be Solved by the Invention" However, in this case, the vertical edge lines are clearly projected with shading at the boundary, and detection of these lines is relatively easy, but the horizontal edge lines are between the left and right rice rows. Due to such factors as gaps, each individual rice plant was obtained as image data, making it difficult to detect horizontal edge lines.

[Means for Solving the Problems] Accordingly, the present invention scans an imaged field screen to detect boundary edge candidate points between crop areas and non-crop areas, and also performs Howe transformation on these candidate points. The crop area edge detection means is configured to identify two intersecting approximate edge lines, which detects a specific value of one of the approximate edge lines based on an example edge candidate point having a high degree of sharpness during the Howe transformation process; , a specific range of the other edge line is established based on the converted value of the corner candidate point at one side edge and the specific value, and the specific value is detected.

"Function" According to the present invention, when processing polar coordinates on the p-θ plane in Howe transformation, one edge line with high sharpness is identified, so that the polar coordinates are processed on the p-θ plane. From the corner candidate point represented by one sine curve above and the corner angle of the two intersecting edge lines, it is possible to roughly identify the other edge line (the intersection of the curves on the ρ-θ plane) that passes through the corner point. The actual processing work of performing Howe transformation on these edge candidate points and identifying edge lines is performed by Dog 1.
] The processing time can be shortened by simplifying the process.

"Example" Hereinafter, an example of the present invention will be described in detail based on the drawings. 1st
The figure is a control circuit diagram, Figure 2 is an overall side view of the combine harvester, and Figure 3 is a plan view of the same.
) is mounted on the truck frame (3), (4) is equipped with an axial flow type screw-type handling barrel (5) and a sorting mechanism (6), and a threshing unit (11) is mounted on the machine base (11). 7) is a grain tank that stores the grains of the threshing section (4) which are taken out via the grain lifting cylinder (8); and (9) is a grain tank that can be raised and lowered via a hydraulic cylinder (10) located in front of the lower part of the threshing section (4). (11) is a driver's seat that is equipped with a driver's seat (12) and a driving operation part (13) and is fixed in front of the grain tank (7); (14) is a driver's seat that is installed in the grain tank (7); an engine room (16) provided at the rear of 7) in which the engine (15) is installed;
is a grain delivery auger that takes out grains from the grain tank (7).

The reaping section (9) is connected to a grain reaping header (7) for receiving uncut grain culms, and to a substantially rear shaft of the header (17) to send the reaped husks to the threshing section (4). It is composed of a supply chamber (18) which is a feeder house for supplying grain, and also includes a reel (19) for scraping in uncut grain culms, a reciprocating drive type cylinder 1 and a second cutting blade (201F21), and a culm scraping auger ( 22) and a forced grain culm pulling device (23) in the grain header (17), and a supply chain conveyor (24) in which the reaped husk culm taken into the header (17) is installed in the supply chamber (18). ) to the threshing section (4) for threshing.

Additionally, a camera support (
25), and a CCD (solid-state image sensor) color camera (26) for capturing an image of the field in front of the travel direction is installed at the upper end of the support column (25) to display an image of the field on an imaging screen such as an x-y plane. It is configured to obtain information.

Furthermore, as shown in FIG. 1, RGB signals (red (R), green (G)
) and blue (B)) to separate and extract seedlings from the field scene 2
A digitization circuit (27) and a rice cultivation area (N) and a non-cultivation area (M) represented in this binarized image on the x-y plane.
Vertical and horizontal edge lines (a, ) (
C2) is transformed into a sine curve (ρ = x cos θ + y sin θ) on the ρ-θ plane of polar coordinates, and the linear approximation edge line [All (A2) of this edge line (a,) (C2) is obtained, Based on the image processing circuit (28) that calculates the intersection (C2) of these lines (A, HA21) and the output signals of the line (A, ) (A2) and the intersection (C2) detected by the circuit (28), A control circuit (31), which is a central processing unit that operates the side clutch solenoid (291 (301) and performs automatic steering of the aircraft, which controls the left and right side clutches, and a steering amount sensor (32) that detects the amount of steering of the aircraft. and a traveling transmission mechanism (33) that stops the aircraft from traveling and changes the traveling speed.

This embodiment is configured as described above, and when the combine harvester is operated automatically in a field as shown in FIGS. 5 and 6, the rice cultivation area [ This method accurately detects the corner (N), changes the direction and attitude of the aircraft, and allows the aircraft to travel appropriately along the edge line in this area. A means for detecting (C,) by image processing will be explained.

When the digitized image is now captured, a binarization process is performed in which the RGB signal level satisfies each rice extraction condition as "1" and otherwise as "0", and then as shown in Figure 7. As shown in , the direction from the top of the screen parallel to the y-axis from the non-cultivated area (M) on the right side to the cultivated area (N) on the left side is as shown by the solid arrow (scanning is performed, and continuous [I
Search for a change point that is a differential point that changes to J. In other words, this point of change becomes the boundary point of area +M+ (N),
By tracing the screen from top to bottom in a fixed order, a large number of vertical edge candidate points (xn, yn), which are changing points, are found on the boundary vertical edge line (a,) with the area (Ml[N). is extracted. and the cultivation area (N
When extracting the lowest corner candidate point (xc, yj) of the vertical edge line (a,) which is the corner (CI) of l,
From the left side of the screen, scan the lower area parallel to the y-axis (the area (N) above Ml) as shown by the dashed arrow, and see how many books are displayed.
The point of change from "0" to "1" is searched for in the same manner as described above, and a large number of horizontal edge candidate points (Xm, yml are extracted on the boundary horizontal edge line (a2) with the area (Ml (Nl)).

In addition, when scanning in this way, the area where rice is present is known in advance, so when scanning starts from the edge of the screen where rice does not exist and extracting a change point, at that point the next column of scanning is started. By repeating these processing operations, it is possible to speed up the search. In this case, how many books change from consecutive "0" to consecutive "1" and when it is confirmed that there is no change in this "1" such as a certain number of consecutive "1"s, the next It moves to row scanning, and is not affected by non-cultivation areas (even if uncut culms or foreign objects are extracted during M1 scanning and the value changes temporarily to "1"). It was set up like this.

And each of these candidate points (xn, yn), (xc, yj
, (Xm, ym) are extracted, each edge line (
Howe transform processing is performed to obtain a linear approximation of (a,) (a2), and each candidate point (xn, y,), (X
c, y, ), (x., ym) i (, ρ
It is converted into a geometric parameter of polar coordinates expressed by a sine curve of ρ=xcosθ+ysinθ on the −θ plane.

That is, between the x-y plane and the p-e plane,
One point (x+, y+) on the straight line "2" on the y plane is ρ-
On the θ plane, the sine curve ρ=X + CO8θ+y, sin
Converted to θ, straight line ρl on the x-y plane :XC
O3θ, +ysinθ1 is one point on the ρ-O plane (ρ1
, θ1). Therefore, if there are many points on the same straight line on the x-y plane, one intersection point of many sinusoids drawn corresponding to these points on the p-θ plane is By finding this, a straight line on the xy plane can be found. In this case, the more points there are on the same straight line on the x-y plane, the more intersections will be formed by a larger number of curves on the p-〇 plane. As shown in the figure, the edge line (a,
) can be detected relatively sharply, but the horizontal edge line (a2) lacks sharpness as each rice husk is obtained as image data due to the distance between the left and right rice rows. Edge candidate points (X
n, , Yn), (x, , , y, ,), the number of candidate points [xn, yn) on the vertical edge line + aj side usually exceeds the number of candidate points [xn, yn), and on the p-θ plane, the number of candidate points [xn, yn) is the highest. An approximate vertical edge line (A1) can be obtained by finding the intersection point (ρ3, θ1) where the two intersect as a specific value.

Therefore, on the ρ-θ plane, each candidate point (xn, ynl, (x
m,y. 1. (A group of curves corresponding to the number of xc, yj is obtained, and then the coordinate point where the most curves intersect, that is, the intersection point (ρ1, θ1) among this group of curves is searched.
The intersection search window (W,) separated by a width of △θ1 is −π/
2 to π/2, and find the above intersection (ρ
1, θ,). And corner candidate point fxe,y
When j is represented by a single sinusoidal curve (S) on the ρ-θ plane, this curve (Sl also passes near the intersection (p+, θ1), so the point (Xc, yj is the approximate edge line (A1 ), the curve (S) is on the intersection (ρ1, θ
1), and when determining the horizontal approximate edge line (A2), the specific value of the intersection point (ρ2, θ2) is a corner angle of approximately 90 degrees (π/2) with respect to the intersection point (p+, θ1). The sinusoidal curve of the separated position (presumed to be on or near Sl, and this intersection on the ρ-θ plane (
ρ2, C2) At the time of search, the θ position is θ, +-rr
, /2 and the p position follows the curve fs) (Z
), the intersection (p2) corresponding to the horizontal approximate edge line (A2) on the , C2) can be easily found. Therefore, when searching for this intersection (ρ2, C2), Δθ2 (Δθ
2. Move the intersection search window (W2) divided by the width of Δθ heat within zone (Z) 03 to efficiently narrow down the search width for the intersection (p2, C2) within a certain range. This makes it possible to reliably search for the intersection (ρ2, C2).

And as shown in Fig. 8 (, these intersections (pl, θI),
When (ρ2, C2) is searched, ρ+”XCO3θ,
+ysinθ8, p2=xcosθ2+ysinθ2
Two approximate edge lines (A1) (A2) are specified in the x-y plane from the relational expression, and these lines (AI)
The intersection obtained from (A2) becomes the approximate edge line corner point (C2) in the cultivation area (Nl), and the corner point (
After detection of C2), the clutch solenoid (29) (30) is operated as appropriate based on this detected value,
This controls the steering of the aircraft.

"Effects of the Invention" As is clear from the above embodiments, the present invention scans an imaged field screen to
), boundary edge candidate points (xn, yN), (Xm, y
m) and detect these candidate points (Xn, 3'n)
, (xm, ym) intersect by Howe transform 2
In the crop area edge detecting means configured to identify the approximate edge line (A, l (A21) of the book, one edge candidate point (X
A specific value (pl, θI) of one approximate edge line (A1) is detected based on the approximate edge line (A1), and a converted value fsl of the corner candidate point (xc, yel) of this one side edge and the specific value (ρ1, pl ) based on the other edge line (A2
), and its specific value (ρ2, C2) is detected. Therefore, the approximate edge line can be identified quickly and reliably by Howe-transforming each edge candidate point, improving processing capacity. It has remarkable effects, such as being able to

[Brief explanation of drawings]

Fig. 1 is a control circuit diagram, Fig. 2 is an overall side view of the combine, Fig. 3 is a plan view thereof, Fig. 4 is a flowchart, Fig. 5 is an explanatory diagram of work, Fig. 6 is an explanatory diagram of an imaging screen, FIG. 7 is an explanatory diagram of edge candidate points, FIG. 8 is an explanatory diagram of Howe transform coordinates, and FIG. 9 is an explanatory diagram of approximate edge lines. (A,) (A2)... Approximate edge line (M)
・・・ Non-crop area

Claims (1)

[Claims] A crop that scans the imaged field screen to detect boundary edge candidate points between crop areas and non-crop areas, and also identifies two intersecting approximate edge lines by performing Howe transformation on these candidate points. The area edge detection means detects a specific value of one approximate edge line based on the edge candidate point on one side having a high degree of sharpness during the Howe transformation process, and also detects a converted value of the corner candidate point at the edge of the one side. An image processing device for a mobile work machine, characterized in that the specific range of the other edge line is established based on the specific value and the specific value is detected.