JPS6082904A - Contour detecting method of object to be inspected - Google Patents

Abstract

PURPOSE:To execute exactly a measurement by irradiating successively an illuminating light to an object to be inspected from plural directions, adding successively a video signal of an obtained image pickup output, and detecting a contour of the object to be inspected by a variation point of its signal. CONSTITUTION:A video signal of an image pickup outout obtained from a television camera CAM for executing an image pickup to an object 1 from a prescribed direction is compared with a reference voltage and binary-coded at every picture element by a comparator CP operated in accordance with a control signal. This binary-coded signal is stored successively in a video memory VDM in accordance with the same control signal. When one cycle of a control signal CS is ended, an operating part PRS reads out the contents of the video memory VDM, and a point where an added value is varied exceeding a prescribed set value is extracted with regard to the horizontal direction and the vertical direction of each address. This variation point is set, for instance, as ''1'' of a logical value, a no-variation point is set, for instance, as ''0'' of the same, and they are stored in a memory MEM. Thereafter, the operating part PRS sends them out as an output data DO.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an improvement in a method for detecting the entire outline of an object to be examined.

[Prior art]

For the purpose of measuring the dimensions of a test object or detecting the position of the test object during automatic processing, the entire test object is imaged using a television camera, etc., and the outline of the test object is detected from the video signal obtained. However, in the past, the object to be examined was illuminated with a completely fixed lighting device, so if the object to be examined had a complex shape, shadows could occur. This is a drawback in that it is not possible to accurately detect the outline of the object to be examined.

[Summary of the invention]

The purpose of the present invention is to fundamentally eliminate the drawbacks of the conventional methods, and the object to be examined is sequentially irradiated with illumination light from a plurality of directions, the object to be examined is imaged from a fixed direction, and the illumination light is It is an extremely effective method that sequentially adds all the video signals of the imaging output obtained in response to sequential irradiation, extracts all the changing points of the added signals, and detects the outline of the object to be examined based on the changing points. , which provides a method for detecting the contour of an object to be examined.

〔Example〕

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to figures showing examples, but first a detailed description of the present invention will be given.

Figure 1 shows an object to be tested, such as a transistor, which has a cubic body with legs protruding in four directions, and shows a case in which object 1 is irradiated with all illumination light sequentially from illumination directions 2 to 9 indicated by arrows. If all images are taken from the visual direction with respect to the paper surface, the perspective view of FIG.
) to (h) are obtained.

Therefore, by adding up the images in FIGS. 2(a) to 2(h), the image shown in FIG. 3 is obtained, and the entire image of the object 1 shown in FIG. 1 can be completely reproduced.

However, in Fig. 3, non-existent contour lines 10 to 23, etc. formed by the shaded areas in Fig. 2 (a) to (h) are generated, but since these do not appear frequently, they cannot be calculated by addition. The degree of impact is reduced.

Figure 4 shows a cubic body connected to an electronic component 24 such as an integrated circuit with lead wires on both sides, and connected to an object to be detected by soldering or bonding all other conductors. 25 to 30, the illumination light is sequentially irradiated from the illumination directions 31 and 32, and the imaging is performed from the visual direction force on the paper surface, and the imaging sensitivity is according to the amount of reflected light from the gold objects 25 to 30. If it is determined, the lighting direction 3
1°32, the images shown in FIGS. 5(a) and 5(b) are obtained.

Therefore, if we add up all the images in FIGS. 5(a) and (b), we will get the image shown in FIG. 6.The object 25 shown in FIG.
~30 total images are completely reproduced.

In other words, if the outline is detected from the overall image obtained using the above principle, the influence of shadows caused by a specific illumination direction will be completely removed, and the entire outline of the object as seen from the imaging direction will be accurately detected. .

FIG. 7 is a perspective view showing an embodiment of means used to sequentially irradiate illumination light from a plurality of directions based on the above-mentioned principle. A plurality of projectors 33a to 33n are arranged in an annular manner, and the direction of light projection is focused on the object 1, and each of the projectors 33a to 3
3n are to be lit sequentially.

In addition, in the same figure (b), the entire light projector 35 is lowered below the rotating body 34b fixed to the rotating shaft 34a, and its light projection direction is aligned with the object 1 directly below the rotating shaft 34a, and the rotating body 34b All of the floodlights 35 are blinked in accordance with the rotation of the .

On the other hand, the same figure (C) shows a case where an annular illumination lamp 36 is used, and a rotating circle with a notch 37b partially extending from the outer periphery to the lower end of a rotating shaft 37a passing perpendicularly through the center thereof. The center of the plate 3γC is fixed horizontally, and the rotating shaft 37a
Illumination light is irradiated onto the object 1 under M from each direction through the cutout 37b, and the illumination light 3G is fully blinked in accordance with the rotational direction of the cutout 37b.

FIG. 8 is a block diagram showing an example of the signal system. The video signal of the imaging output obtained from the television camera (hereinafter referred to as camera) CAM which images the object 1 from a fixed direction is controlled by the control unit CNT. The comparator CP, which operates according to the signal, compares it with the reference voltage and converts it into a binary signal for each pixel, and converts this binary signal into a video memo IJVDM according to the same control signal.
This is because the above operations are repeated in response to the control signal cs sent by the control unit CNT to control the blinking of the floodlights 33a to 33n, 35t or the illumination light 36 in FIG. , the binarized signals each time the object 1 is irradiated with illumination light from each direction are sequentially added for each pixel in the video memory VDM having an addition function for each address, and the added value is converted into a decimal number. In other words, the signal after the addition is in the state illustrated in FIG. 1(a).

Moreover, 1-! of the control signal C8! l? - When the cycle is completed and the irradiation of all the illumination directional forces and the like to the object 1 is completed, the control unit CNT or the calculation unit PR8 of an electronic computer etc.
In order to send a control signal to the video memo IJVDM, the calculation unit PR8 reads out the contents of the video memo IJVDM, and the point at which the added value changes beyond a predetermined set value? Extract the horizontal and vertical directions of each address, set this change point as, for example, a logical value of 11", and set all unchanged points as, for example, the same -θ" as a memo IJM.
Store in EM.

In addition, at this time, the image position may change slightly at each imaging time point.
For the purpose of removing this influence, the change points are expressed and extracted by two bits, for example.

Therefore, the memory MEM K stores a signal in the state illustrated in FIG. 9(b) in which the noise component in FIG. 9(a) or the influence other than 0.8 including this has been removed. All arithmetic unit PR3 reads out the output data D.
If it is sent as o, the entire outline of the object 1 can be detected based on this.

Also, video memo! The contents of JVDM are shown on the video monitor V
It is also given to M'H, which makes it possible to monitor the entire imaging situation and addition situation.

However, when using the illumination means shown in FIG. 7(b) or (C), the floodlight 35 or illumination lamp 36 is turned on continuously, the entire rotation angle of the rotating part is detected, and the control shown in FIG. 8 is performed. Part C
NT, and a control signal 'r% may be generated for each part according to this detected force.In addition to the configuration shown in FIG. Various configurations can be selected, such as one that performs irradiation.

Further, the illumination direction may be set in any direction as long as the contour to be detected is clearly visible, and can be set arbitrarily depending on the conditions.

In Fig. 8, a scanning mirror and a 7-otodiode array may be used instead of the camera CAM, and the image No. 43 is converted into a multi-bit digital signal by an analog-to-digital converter. The same effect can be achieved even if the data is stored in the buffer memory and then sequentially added.If an imaging unit such as a camera CAM that has a video signal accumulation mode is used, the addition process on the circuit can be omitted entirely. Various modifications can be made freely.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, the influence of object shadows is eliminated, and noise components are removed by adding video signals obtained by capturing multiple images. Detection is accurate, and remarkable effects can be obtained for each line of use, such as object dimension measurement and position detection.

[Brief explanation of the drawing]

1 to 6 are perspective views showing the principle of the present invention, FIG. 7 and subsequent figures show embodiments of the invention, FIG. 7 is a perspective view of the illumination means, FIG. 8 is a block diagram of the signal system, FIG. 9 is a diagram illustrating the contents of each memory in FIG. 8. 1.25~30...Object (test object), 2~9,3
1.32...Illumination direction, 33a to 33n. 35... Floodlight, 34a, 37a... Rotating shaft, 34b... Rotating body, 36... Lighting lamp,
37c...Rotating disk, CAM...Camera (TV camera), CP...Comparator, V D
M: Video memory, PR8: Arithmetic unit, MEM: Memory, CNT: Control unit. Patent applicant Hitachi Electronics Co., Ltd. Agent Masaki Yamakawa (and one other person) Figure 2 Figure 2 (() (d) (e) (f) Figure 3 Figure 4 (o> (b) Figure 6 B

Claims (1)

[Claims] Illumination light is sequentially irradiated onto the test object from multiple directions,
The object to be inspected is imaged from all fixed directions, and the video signals of the imaging outputs obtained in response to the sequential irradiation of the illumination light are sequentially added, all the changing points of the added signals are extracted, and the changing points are A method for detecting an outline of an object to be examined, characterized in that the outline of the object to be examined is detected.