JPS61294985A - Detecting device for rotation slippage - Google Patents

Abstract

PURPOSE:To detect the rotation slippage volume of an input pattern against a reference pattern at a high speed and to rapidly and easily recognize an object by providing a pattern generating means, a data taking out means, a storing means and a means which discriminates the rotation slippage volume of the input pattern. CONSTITUTION:When a reference model in a state that no rotation slippage is occurred is image-picked up by a television camera 1, a binary-coded process is performed on the first odf field of a video siganl VDi generating th reference pattern P0 and it is sent to a black picture element detecting circuit 5 detecting a black picture element and with calculating the number of the black picture elements at a picture element counter 6 and the sum of horizontal load at an adder 7, the result is stored at a RAM 14. When the object to be recognized is image-pickuped by the television camera 1, an input pattern Pi having the rotation slippage of angle theta is generated and is processed as previously. Stored data regarding the input pattern Pi are checked in order with the reference data of each of reference patterns P0-Pn and a corresponding angle is detectd as the rotation slippage volume of the input pattern Pi with searching the refer ence data having the highest coincidence degree.

Description

[Detailed Description of the Invention] <Technical Field of the Invention> This invention obtains an input pattern by imaging a stationary or moving object to be recognized, and compares this input pattern with a standard pattern to identify the object to be recognized. In relation to visual recognition technology for subsequent recognition, in particular, the present invention provides a novel rotational deviation detection device that rapidly detects the amount of rotational deviation of an input pattern with respect to a standard pattern prior to pattern matching.

<Summary of the Invention> The present invention calculates the amount of rotational shift of the input pattern by obtaining input data regarding pattern constituent pixels for each horizontal scanning line for an input pattern, and then comparing this with standard data obtained in the same manner in advance. This enables high-speed detection of the amount of rotational deviation of the input pattern at video rates.

<Background of the Invention> In general, a two-dimensional visual recognition device recognizes an object by superimposing an input pattern and a standard pattern on an image, detecting the degree of superimposition and matching between the two patterns.

Therefore, when performing this pattern matching, it is necessary to accurately align both patterns. Conventionally, an XY stage or the like is used to position the object to be recognized at a predetermined stopping position, and then the object is imaged with a television camera to determine the input pattern. This input pattern is then compared with a standard pattern. However, in the case of this type of system, since a mechanism for positioning the object to be recognized is required, the overall structure of the visual recognition device becomes complicated, and there are disadvantages such as the time required for pattern matching corresponding to the time required for the positioning operation.

Therefore, in recent years, the amount of positional deviation and rotational deviation of the input pattern with respect to the standard pattern is detected as data, the deviation between the input patterns is corrected based on this detected data, and then pattern matching for object recognition is performed. A method was developed.

However, in the past, in order to specifically detect rotational deviation of an input pattern, the contour of the input pattern was first extracted, and then the moment of the input pattern was determined from this contour data.
This moment is used to calculate the amount of rotational deviation of the input pattern, which takes an extremely long calculation time, making it difficult to achieve high-speed processing.

<Object of the Invention> The present invention provides a rotational deviation detection device that is capable of detecting the amount of rotational deviation of an input pattern with respect to a standard pattern at high speed, thereby enabling quick and easy object recognition even if the input pattern is tilted. The purpose is to

<Structure and Effects of the Invention> In order to achieve the above object, the rotational deviation detection device of the present invention includes a pattern generation means for binarizing an object image to generate a standard pattern and an input pattern, and a pattern generation means for generating a standard pattern and an input pattern. data retrieval means for retrieving data regarding the pattern composing pixels for each horizontal scanning line; and for each standard pattern given a rotational shift of a different angle, storing data regarding the pattern composing pixels obtained in advance as standard data for each angle. storage means,
The apparatus further includes a determining means for comparing the input data regarding the pattern constituent pixels obtained for the input pattern with the standard data to determine the amount of rotational deviation of the input pattern.

According to this invention, since it is possible to correct rotational deviation between patterns on the data, there is no need to position the object to be recognized at a predetermined stop position (no special positioning mechanism is required, and the entire device In addition, the time required for positioning operations can be saved, and the efficiency of object recognition processing can be improved.

In addition, the amount of rotational deviation of the input pattern can be detected using a counter or other simple hardware configuration, or even with a video radar, instead of using complex arithmetic processing using moments, increasing the speed and efficiency of arithmetic processing. Improvement can be achieved.

Furthermore, data that is not easily affected by image noise, such as the horizontal weighted sum average, can be used as rotational deviation detection data, so even if the input pattern contains image noise, the noise part is not part of the pattern. This reduces the possibility of misperception that there is a rotational deviation, and makes it possible to accurately detect and correct the amount of rotational deviation, achieving the remarkable effects of achieving the purpose of the invention.

<Description of Embodiments> FIG. 1 shows an example of a circuit configuration of a rotational deviation detection device according to the present invention. In the illustrated example, the television camera 1 images a stationary or moving object 2, for example from above, and sends an image output (shown in FIG. 2(1)) related to interlaced scanning to the synchronization separation circuit 3. The synchronization separation circuit 3 extracts a horizontal synchronization signal HD, a vertical synchronization signal VD, an odd field signal OD (shown in FIG. 2 (2)), and a clock signal CK from the image output.
(shown in Figure 2 (4)), etc., and separate the video signal VDi.
is output to the binarization circuit 4. The binarization circuit 4 is shown in FIG.
As shown in 3), a certain threshold level TH is set for the video signal VDi, and the odd fields of the video signal VDi are converted into black and white binarized to form a binary pattern and output.

FIG. 3 shows an image of a binary pattern, and FIG. 3 (1) shows an elliptical standard pattern P. , Figure 3 (2)
denote the same elliptical input pattern P, respectively. In the figure, the shaded areas are black pixels forming the pattern, and the areas other than the shaded areas are white pixels corresponding to the background. In this embodiment, each pattern is assembled from a pixel range of 256 bits in the vertical and horizontal directions, and in the case of the input pattern P shown in FIG. ing.

A black pixel detection circuit 5 is connected to the binarization circuit 4,
Furthermore, a pixel counter 6 is connected to the black pixel detection circuit 5. The black pixel detection circuit 5 detects the black pixels (shaded areas in FIG. 3) constituting each pattern, and the pixel counter 6 counts the output (number of black pixels) of the black pixel detection circuit 5. Further, an adder 7 and a buffer memory 8 are connected to the output side of the black pixel detection circuit 5, and the adder 7 and buffer register 8 store the counted value of the horizontal counter 9 in the hard drive at each black pixel detection timing. At the same time, the added value for one horizontal scanning line (this is referred to as a "horizontal weighted sum") is accumulated and held in the buffer memory 8.

In FIG. 1, the horizontal counter 9 and the vertical counter 10 are used to address the pixel position of the pattern when the pattern is taken in. It's Nishide. Further, the gate circuit 11 is controlled to open and close by an odd field signal OD, and supplies a clock signal CK to the horizontal counter 9 and the vertical counter 10, respectively.

The count data (number of black pixels) of the pixel counter 6 and the cumulative data (horizontal load sum) of the buffer register 8 are inputted to I/O (Input/0utpu) at the end of each horizontal scan.
t) CPU (Central P
processing unit) 13,
The CPU 13 divides the horizontal weight sum by the number of black pixels to obtain an average horizontal weight sum, and stores this calculation result in RAM (Random
Access Memory) 14 and displays it on the display unit 15, and also executes a series of calculations and processes related to detection of rotational deviation of the input pattern P.

In addition, in the figure, F ROM (Programmable
ReadOnly Memory) 16 stores a series of programs for detecting and correcting rotational deviations, and also
The AM 14 has an area for storing various data as well as a work area for executing various processes. Further, the gate circuit 17 is a circuit for generating an interrupt signal INT to the CPU 13, and the OR circuit 18 is a circuit for resetting the pixel counter 6.

Fig. 4 shows a standard pattern P0 with no rotational deviation, and Fig. 5 shows a standard pattern P given a rotational deviation of angle θ1.
FIG. 6 shows a standard pattern P2 given a rotational deviation of a larger angle θ2, and FIG. 7 shows an input pattern P having a rotational deviation of a certain angle θ. In each figure, X indicates a horizontal address and Y indicates a vertical address. Also, the histogram shown for each time H8+ HI IH,,H
indicates the size of the horizontal weighted sum average for each horizontal scanning line, and the area diagram of the RAM 14 indicates the storage range (shaded area in the figure) of each horizontal weighted sum average. According to the figure, it can be seen that as the rotational deviation of the pattern increases, the shape of the histogram changes and the data storage range of the RAM 14 becomes narrower.

Therefore, prior to object recognition processing, first the standard model with no rotational deviation, and then each standard model given a rotational deviation of a different angle, use the circuit shown in Fig. 1 to calculate its standard pattern P O+ P I+ P 21...
・Create standard data for detecting rotational deviation.

When the standard model with no rotational shift is imaged by the television camera 1, a binarization process is performed on the first odd field of the video signal VD, and the standard pattern P0 shown in FIG. 4 is generated. This standard pattern P0 is
The signal is sent to the black pixel detection circuit 5, where black pixels are detected, and the pixel counter 6 calculates the number of black pixels, and the adder 7 calculates the horizontal weighted sum. Then, an interrupt signal INT is generated to the CPU 13 at each time timing of the horizontal synchronization signal HD in the odd field, and the CPU 13 takes in the number of black pixels in the pixel counter 6 and the horizontal weight sum in the buffer register 8 during this horizontal synchronization period. , calculate the horizontal weighted sum average, and use the result as R
Store in AM14. This calculation of the horizontal weighted sum average is performed line by line across all horizontal scanning lines, and in this case, each calculated value is the angle θ of the RAM 17. The information is sequentially stored in the area of the address corresponding to (see Figure 4).

5 and 6 show angles θ7. It shows standard patterns P, P2 given a rotational shift of θ2, and after calculating the horizontal weighted sum average for each of these patterns P+ and Pz using the same procedure as above, each data is stored in the RAM 14. The angles θ and θ2 are stored in the address areas corresponding to the angles θ and θ2.

Below, further different angles θ5. θ4. ..., standard pattern P3. with a rotational shift of θ1. P4.・・・・・・
, P7 (not shown), the horizontal weight sum average is calculated using the same procedure, and each data is stored in the RAM 14, thereby completing the generation of standard data for detecting rotational deviation.

Next, when the object to be recognized is imaged by the television camera 1, a binarization process is performed on the first odd field of the video signal VDi, and an input pattern P having a rotational shift of an angle θ as shown in FIG. generated. This input pattern pt is sent to the black pixel detection circuit 5, where black pixels are detected, the pixel counter 6 calculates the number of black pixels, and the adder 7 calculates the horizontal weight sum.

Then, at each time timing of the horizontal synchronization signal HD in the odd field, an interrupt signal INT is issued to the CPU 13, and during this horizontal synchronization period, the CPU 13 takes in the number of black pixels in the pixel counter 6 and the horizontal weight sum in the buffer register 8, and The horizontal weighted sum average is calculated and the result is stored in a predetermined area of the RAM 14. When this calculation of the horizontal weighted sum average for each horizontal scanning line is executed over all the horizontal scanning lines, the CPU 13 then stores the data stored in the RAM 14 regarding this input pattern Pi in the period of the subsequent even field. The standard patterns PO+PI, PZ+, . Thereafter, the CPU 13 corrects the rotational deviation of the input pattern P based on the detected angle, and starts pattern matching processing for object recognition using the corrected pattern.

In the above embodiment, the amount of rotational deviation of the input pattern P is obtained using the horizontal weighted sum average, but instead of this horizontal weighted sum average, the rotational deviation can also be detected using the number of black pixels for each horizontal scanning line. Processing is possible.

[Brief explanation of the drawing]

Fig. 1 is a circuit block diagram of a rotational deviation detection device according to the present invention, Fig. 2 is a time chart showing signal waveforms of the circuit of Fig. 1, and Fig. 3 (11 (21 is a diagram showing a standard pattern and an input pattern) , FIGS. 4 to 7 are diagrams showing the principle of detecting the rotational deviation of the input pattern with respect to the standard pattern. 1...Television camera 4...Binarization circuit 5
...Black pixel detection circuit 6...Pixel counter 7...Adder 8...
・・Buffer register 9・・Horizontal counter 13・・・・CPU 14
...RAM patent applicant Shigeishi Electric Co., Ltd. Agent
Patent Attorney Yu Mitsuru Suzuki'd 4 m 1 apt, re (a pass, x, zxt,
r, re size 5)" 1 rotation, j4 (accumulation principle 1 hos1ri'Aj 61.ffi r84z, tnMLc,
tf? , ct * rr= t + fl”IF '7
] ¥1 rtbt*krt, /r, J'y:
&, ttts pass

Claims (4)

[Claims] (1) a pattern generation means for binarizing an object image to generate a standard pattern and an input pattern; a data extraction means for extracting data regarding pattern constituent pixels for each horizontal scanning line for the standard pattern and input pattern; storage means for storing data regarding the pattern constituent pixels obtained in advance for each angle as standard data for each standard pattern given a rotational shift; and storing input data regarding the pattern constituent pixels obtained for the input pattern as standard data. 1. A rotational deviation detection device comprising: determining means for determining the amount of rotational deviation of an input pattern by comparing it with the input pattern. (2) The rotational deviation detection device according to claim 1, wherein the pattern generation means includes a television camera that images an object, and a binarization circuit that binarizes the object image. (3) The rotational deviation detection device according to claim 1, wherein the standard data and the input data are horizontal weighted sum averages of patterns obtained for each horizontal scanning line. (4) The rotational shift detection device according to claim 1, wherein the standard data and input data are the number of pixels forming a pattern determined for each horizontal scanning line.