JPH03214695A - Image recognizer of electronic parts mounting equipment - 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 Fields] The present invention is applicable to, for example, the external shape, position,
The present invention relates to an image recognition device for an electronic component mounting machine that recognizes inclinations, leads, etc.

[Conventional technology] In the assembly process of electronic components (hereinafter simply referred to as components), components are transferred using a component suction machine and placed in a predetermined position.

By the way, in an electronic component mounting machine (hereinafter simply referred to as a mounting machine) equipped with an image recognition device for correcting the positional shift and tilt of the picked-up component and inspecting the external shape of the component, the image input section of the image recognition device is In order to maintain the reliability of the mounting machine, it is important to detect foreign objects such as dust entering the component field of view or components that have fallen into the component field of view due to a suction error, and detecting deterioration of the lighting that illuminates the components.

Conventionally, as shown in Figure 4, regarding foreign objects within the field of view of parts, regular visual inspections were carried out to determine whether a normal part was defective due to reasons such as dust adhering to the part leads during external inspection of the part. Something was happening.

Furthermore, deterioration of lighting has been detected using a sensor that monitors the amount of light from the lighting when it no longer meets a predetermined amount of light.

[Problems to be Solved by the Invention] However, with conventional technology, foreign objects within the field of view of the component are detected only after the performance of the mounter has deteriorated due to foreign object contamination, which reduces the reliability of the mounter. At the same time, a problem has arisen in that normal, expensive parts are determined to be defective based on external shape recognition and are discarded. Further, regarding the detection of deterioration of lighting, the use of expensive sensors increases costs, which is contrary to the aim of providing the mounting machine users at the lowest possible cost.

The present invention solves the problems of the prior art described above.
The present invention provides an image recognition device for an electronic component mounting machine that can prevent a decrease in the reliability of the mounting machine and prevent the disposal of normal and expensive components.

[Means for Solving the Problems] In order to achieve the above object, the present invention has the following configuration. That is, the present invention provides an electronic component mounting machine equipped with an image recognition device that recognizes at least the position and external shape of an electronic component. The brightness information on a plurality of predetermined straight lines is compared with the brightness information on a plurality of predetermined straight lines in a pre-stored image of a normal state without electronic components, and the difference between these two kinds of brightness information is calculated. According to the present invention, there is provided an image recognition device for an electronic component mounting machine, characterized in that the image recognition device includes means for detecting foreign objects and illumination deterioration within the field of view of the component.

[Operation] According to the above configuration of the present invention, since an image captured before the component enters the component field of view during the mounting operation is used, foreign matter that has entered the component field of view is detected before the component is recognized. This can be detected, reducing the reliability of the mounting machine and preventing the disposal of normal, expensive parts. In addition, since the information to be compared is limited to the luminance information on predetermined multiple straight lines of the image,
Does not require processing time that interferes with implementation operations.

Furthermore, since luminance information in a normal lighting state is used as the source information for comparison, detection of lighting deterioration can be performed at the same time, making it possible to eliminate the need for a sensor for detecting lighting deterioration.

[Example 4] Hereinafter, the present invention will be explained in more detail using an example. Note that the present invention is not limited to the following example.

FIG. 1 is a flowchart of a method for detecting foreign objects and illumination deterioration using an image recognition device of a mounting machine according to an embodiment of the present invention, and FIG. 2 is a model diagram of an image recognition device and an image input unit according to an embodiment of the present invention. , FIG. 3 is a diagram showing an example of brightness information in an embodiment of the present invention.

In FIG. 2, 6 is an image recognition device, 7 is a TV camera in the image input section, 8 is a lens in the image input section, 9 is a nozzle for picking up parts, 10 is a part, and 11 is a device for obtaining an image of the shadow of the part. 12 is a foreign object.

The operation of one embodiment of the present invention will be described below in accordance with the flowchart of FIG. 1 and with reference to FIG. 3.

As preprocessing, an image of a component field of view in which no component is shown in a normal state is captured, luminance information on a plurality of predetermined straight lines is calculated, and stored in the image recognition device.

In this embodiment, the plurality of straight lines are horizontal to the image, and their positions are arranged in a rectangular shape with intervals shorter than the shortest dimension of the smallest component handled by the mounting machine of this embodiment. Furthermore, the brightness information is the sum of brightness at each position on each straight line. An example is shown in FIG. 3. 14 is 1 of multiple straight lines
A straight line is shown, and a graph of the brightness position on this straight line is shown on the right. Under normal conditions, the brightness is approximately constant at each position on the straight line.

In the operation of detecting foreign objects and illumination deterioration of the image recognition device of the mounting machine of this embodiment, first, as shown in FIG. It begins by importing the image into an image recognition device.

Next, as shown in FIG. 1 and 2, the captured image is processed within the image recognition device to calculate brightness information on a plurality of predetermined straight lines. FIG. 3 15 shows an example in which a foreign object has entered the field of view of the component, and FIG. 3 16 shows an example in which the illumination has deteriorated.

Next, as shown in FIG. 1, the stored brightness information and the calculated brightness information are compared on corresponding straight lines. In the comparison, even under normal conditions, it is not possible to obtain luminance information that is exactly the same as the actually stored luminance information, so the comparison is performed within a preset tolerance range. In the example of FIG. 3, for example, the stored brightness information is 82000.
So, assuming that the tolerance range is 5%, there is a difference of just over 9% in example 15, and a difference of about 45% in example 16, so the difference in brightness information exceeds the tolerance range, so 15, 16 Both cases will be detected as foreign objects and illumination deterioration. If the difference in brightness information is within the allowable range, it is determined to be normal as shown in FIG. 14.

The above is the operation of detecting foreign matter and illumination deterioration in this embodiment. In this embodiment, foreign object detection and illumination deterioration detection are not distinguished, but since foreign objects generally exist in a part of the component field of view, the brightness information of the decimal straight line is different, and illumination deterioration generally affects the entire component field of view. Therefore, the brightness information on many straight lines is different, and if this is added to the determination, it is possible to distinguish between foreign objects and illumination deterioration.

[Effects of the Invention] According to the image recognition device of the mounter of the present invention, foreign matter that has entered the field of view of the component is detected before the component is recognized during the mounting operation, which reduces the reliability of the mounter and prevents it from operating normally. The effect of preventing the disposal of expensive parts can be achieved.

At the same time, it is also possible to detect deterioration of lighting, eliminating the need to prepare a special sensor for detecting deterioration of lighting.
It is also possible to achieve the effect of reducing the cost of the mounting machine.

[Brief explanation of drawings]

FIG. 1 is a flowchart of a foreign object/illumination deterioration detection method using an image recognition device of an electronic component mounting machine according to the present invention, and FIG. 2 is a model diagram of an image recognition device and an image input section in a mounting machine according to an embodiment of the present invention. , FIG. 3 is a diagram showing an example of brightness information in the same embodiment, and FIG. 4 is an example of a component image when foreign matter is mixed in in the conventional technique. 6... Image recognition device, 7, 8... Image input section, 9.
...Parts suction nozzle, 10...Parts, 11...Lighting, 12...Foreign object. 3rd luminance luminance luminance map coordinates

Claims (1)

[Claims] (1) In an electronic component mounting machine equipped with an image recognition device that recognizes at least the position and outline of an electronic component, a predetermined portion of an image captured during mounting operation before the electronic component enters the field of view of the image input unit. The brightness information on a plurality of straight lines is compared with the brightness information on a predetermined plurality of straight lines in a pre-stored image of a normal state without electronic components, and based on the difference between these two types of brightness information, An image recognition device for an electronic component mounting machine, comprising means for detecting foreign objects and illumination deterioration within a field of view of a component.