JPH0656295B2 - Image processing method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an image processing method for detecting the position of an object using a visual sensor.

(Prior Art) Conventionally, it is known to detect the position of an object based on image data obtained by imaging the object with a visual sensor.

The image data is written in a data table with an address corresponding to the position on the imaging screen of the visual sensor, and the image data of the object is searched from the image data,
The position of the object is detected.

(Problems to be Solved by the Invention) In the above, when there is another object, that is, a non-target object around the target object, the video data of the target object and the video data of the non-target object included in the image data are Therefore, it is necessary to perform the shape determination process of each image to determine the image of the object, and this process takes time, and there is a problem that the entire processing time required for position detection becomes long.

In order to solve such a problem, it is an object of the present invention to provide an image processing method capable of quickly performing determination processing of an object by removing image data of the non-object in advance as much as possible. To do.

(Means for Solving the Problems) In order to achieve the above object, the present invention binarizes image data obtained by capturing an image of an object with a visual sensor and writes the image data in a data table, and based on the image data, In the image processing method for detecting the position of the object, when reading the image data from the data table, the binary signal of the portion where the video exists is set to "1" and the binary signal of the blank portion is set to "0". , "1" is written in a portion where a signal of "1" having an address in contact with the periphery of the image pickup screen of the sensor is written, and in a portion where the signal of "1" is continuously written in the vertical and horizontal directions at the portion.
Is rewritten to "0", the image data in contact with the periphery of the image pickup screen of the sensor is erased from the image data, and the object discrimination process is performed based on this image data. .

(Operation) In general, the visual sensor is roughly positioned with respect to the target object, and the image of the target object is picked up by the sensor so that the image fits within the imaging screen.

In this case, most of the images of the non-objects around the object are partially projected from the imaging screen, and are captured in a state where the protruding part is cut off at the periphery of the screen.

Therefore, if the data of the video that is in contact with the screen is deleted from the image data, the remaining image data will hardly include the image data of the non-target object, and the number of videos that are subject to shape determination will be reduced, and the target object will be reduced. The time required for the determination process of is shortened.

(Embodiment) Referring to FIGS. 1 and 2, a denotes a rear bumper of an automobile, b denotes a pair of mufflers provided under the bumper, and a measurement rod is used as a muffler b in an exhaust gas measurement operation. A visual sensor (1) consisting of a CCD camera when inserted into the
Was used to detect one of the mufflers b.

The sensor (1) is provided at a fixed position behind the work station for loading and setting the vehicle body, and the sensor (1) images a range indicated by an imaginary line in FIG. CC that is configured to be placed on the image plane
The signal from the D element is input to the comparator (3) and the serial shift register (4) via the amplifier (2) as shown in FIG. 3, and the image data composed of the binary serial signal thus obtained is generated. , Write in a data table in the RAM (5) with an address corresponding to the position on the image pickup screen of the sensor (1), and from this image data, the CPU (6)
Thus, the image of the object, that is, the muffler b is discriminated and its position is detected.

In the figure, (7) is a CRTC (Cathode Ray Tube C) that generates an address signal for writing the image data in the RAM (5).
ontroller), (8) indicates a data path, and (9) indicates an address bus.

By the way, the image represented by the image data is, as shown in FIG. 4, in addition to the image B of the muffler b which is the object in the imaging screen (10), the image B ′ of the muffler b adjacent thereto and the rear bumper a. Image B'in this case, in this case image B '
The image A is in contact with the peripheral edge of the screen (10) without being completely inside the screen (10).

Therefore, if the image data in contact with the periphery of the imaging screen (10) is erased from the image data, only the image data B remains in the erased image data as shown in FIG. The shape determination of the other images B ′ and A is not necessary, only by determining the shape of the image of the muffler as the object.

Here, the video data is erased when the image data is read from the table by setting the binary signal of the portion where the video written in the data table is present to "1" and the blank portion thereof to "0". Of the "1" signal having an address in contact with the periphery of the imaging screen (10) and a portion in which the "1" signal is continuously written in the vertical and horizontal directions. This is performed by rewriting the signal to "0", and the erasing process can be performed in an extremely short time compared to the case where the image shape determining process is performed, and the entire processing time required for detecting the position of the target object can be obtained. Can be shortened.

(Effects of the Invention) As described above, according to the present invention, among the images of the non-objects existing around the object, the data of the image which is in contact with the peripheral edge of the imaging screen is extracted from the image data prior to the object discrimination processing. It is erased, the discrimination process can be performed quickly, and the entire processing time can be shortened.

[Brief description of drawings]

FIG. 1 is a plan view showing a positional relationship between a visual sensor and an object, FIG. 2 is a diagram showing an imaging range by the visual sensor, and FIG. 3 is a block diagram of an image processing apparatus used for carrying out the method of the present invention. FIG. 4 is a diagram showing an image based on unprocessed image data, and FIG. 5 is a diagram showing an image based on image data after erasing image data in contact with the periphery of the image pickup screen. (1) ...... Visual sensor (10) ...... Imaging screen B …… Image of object A, B ′ …… Image in contact with the edge of the imaging screen

Claims (1)

[Claims] 1. An image processing method for binarizing image data obtained by picking up an image of an object by a visual sensor, writing the image data in a data table, and detecting the position of the object based on the image data. When the binary signal of the part is set to "1" and the binary signal of the blank part is set to "0", when reading image data from the data table, "1" having an address in contact with the periphery of the image pickup screen of the sensor is set. The signal of "1" is rewritten to "0" at the portion where the signal of "1" is written and the portion where the signal of "1" is continuously written in the vertical and horizontal directions, and is written on the periphery of the image pickup screen of the sensor. An image processing method characterized in that data of a video in contact is erased from image data, and an object discrimination process is performed based on the image data.