JPS63311109A - Method for discriminating crest shape of key by image processing - Google Patents

Abstract

PURPOSE:To efficiently perform highly accurate inspection, in picking up the image of the crest and valley parts of a key by a sensor camera, by setting a window at the position to be measured in a software manner to perform image processing. CONSTITUTION:When the images of the crest part 10 and valley part 9 of a key 1 are picked up by a sensor camera 3 constituted of an image pickup element group, slit like windows 7 are set to the crest part 10 and valley part 9 of the key 1 being the positions to be measured in a software manner to perform image pickup. The shade generated by blocking the windows 7 by the crest part 10 and valley part 9 of the key 1 is caught as the change in the quantity of beam to be taken out as a dimension signal and this signal is subjected to image processing. Therefore, it is unnecessary to move the key 1 and, since the dimensions of the crest part 10 and valley part 9 of the key 1 are simultaneously measured with high accuracy, a large number of the keys can be inspected efficiently with high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention provides a clear key identification method in the automobile industry and the like where it is necessary to manage a huge number and types of keys.

<Prior Art> Conventionally, the following method has been adopted to identify individual keys.

■A projector shines light onto one side of the key's peaks and valleys, and the shadow is magnified by a lens and projected onto a receiver.

■A large number of light receiving elements are arranged at equal intervals on the light receiver.

(2) The height of the peak or valley of the key is determined depending on whether these light-receiving elements receive light or not.

Therefore, the method does not measure the width of the peaks or valleys of the key, but merely determines whether or not they fall within a certain size range. Further, since the discrimination is performed only on one side of the key, there is a problem in that the discrimination accuracy is affected by the play in the key mounting jig, its feeding device, etc.

<Problems to be Solved by the Invention> Therefore, it is desirable to be able to measure the width dimensions of all the peaks and valleys of a key at the same time, and to consider a method that can do this using a simple device. had become an issue.

<Means to solve the problem> As a result of various studies regarding these problems, key (1)
When photographing the peaks (10) and valleys (9) of They developed a method for identifying the shape of a key peak using image processing, which is characterized by identifying the shape of a key peak through processing. ' Setting the window (7) above the image area (8) using software means setting a long slit-like frame on the image area (8) using computer software, although it is invisible to the naked eye. The set state is displayed on the monitor (5). Image processing means capturing and processing the presence of brightness and darkness within the window (7) as a change in light amount.

As a specific example of the window (7), it is programmed to have a thickness corresponding to a slight parallel portion at the positions of the key peaks and valleys, and a width larger than the maximum width of the key peaks.

<Operation> In this method, the light reaching the sensor camera (3) is simultaneously irradiated over the entire measurement range of the key, so there is no need to move the key (1), and the width of the peak or valley of the key can be measured simultaneously and with high precision.

<Example> Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 is a perspective view of a measuring device that makes it possible to implement the key mountain shape identification method of the present invention, and FIG. 2 is a diagram showing the state of an image portion in image processing.

In this device, a key (1) is attached to a jig (2) that allows you to photograph a panoramic view of the mountains and valleys by shining a lighting device (6) from behind, and a sensor camera (3) is attached to the top of the jig (2). ing. This sensor camera (3) has an image section made up of a group of image pickup elements, and photographs the peaks and troughs of the key.

The photographic results are sent to the controller (4). In the controller, a computer software, that is, a program, is used to determine the measurement target position of the image part (8) suitable for image processing, that is, all the peaks (10) and valleys of the key, as shown in FIG. A slit-shaped window (7) having a thickness corresponding to the slightly parallel portion and a width larger than the maximum width of the key crest is set at the position of the portion (9).

With such a device, all slit-like lines (7
The shadows blocked by the peaks or valleys within ) are captured as changes in light intensity, subjected to image processing, and then extracted as a signal for the width of the peaks or valleys of the key.

In general, the most important elements of a key, crests and troughs, are arranged at regular distances in the longitudinal direction.

These peaks or valleys are arranged symmetrically about the key center and have different heights, ie different widths, leaving a slight parallel section at the top or bottom.

The key is identified by a combination of the width dimension of this slightly remaining parallel portion and its longitudinal position. The shape of the slope of the peak or valley connecting these parallel parts is not related to the function of the key. Needless to say, strict dimensional accuracy is required for the key to function as a key. Incidentally, an accuracy of 1/100 mm is now required for the distance from a certain reference position to a slightly parallel part of a peak or valley, and an accuracy of 1/100 ne for the entire width of the peak or valley. Ta. Therefore, these precisions must be taken into consideration when identifying the shape of the key mountain.

Therefore, in the present invention, the following is performed within the controller using the extracted signal.

(2) Search for the key Nα by comparing the key pattern stored in the storage device, that is, the width of the peak or valley of the key and the combination of the longitudinal direction.

■Calculate the width dimensions of all peaks and valleys through arithmetic processing.

(2) Compare the width dimensions of the peaks and valleys with those of the searched key Na to determine quality.

All the results are then displayed on the monitor (5).

<Effects of the Invention> Since the present invention is a key peak shape identification method as described above, it is possible to measure the width of the peak or valley of a key with high precision, which was difficult with conventional methods; , enables high-precision dimensional inspection, and (2) the data obtained through measurement can be effectively used in other processes, such as cutting key ridges or assembling cylinders.

■The device is simpler because there is no need to move the keys during shooting. In addition, it is efficient because simultaneous measurements can be made at all points.

[Brief explanation of drawings]

FIG. 1 is a perspective view of a measuring device that makes it possible to implement the key mountain shape identification method of the present invention, and FIG. 2 is a diagram showing the state of an image portion in image processing. (1) Key (2) Jig (3) Sensor camera (4) Controller (5) Monitor (6) Lighting device (7) Window
(8) Image area (9) Valley of the key (10) Above the peak of the key

Claims (1)

[Claims] 1. When photographing the peaks (10) and valleys (9) of the key (1) with the sensor camera (3), a window (7) is placed on the image area (8) by software in all measurement target areas. A method for identifying the shape of a key mountain by image processing, which is characterized by setting and performing image processing.