Pattern detection method

• US 5,912,985 A
• Filed: 11/07/1995
• Issued: 06/15/1999
• Est. Priority Date: 11/08/1994
• Status: Expired due to Fees

• Current Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Original Assignee: Matsushita Electric Industrial Company Limited (Panasonic Holdings Corporation)
• Inventors: Bozma, Omur, Hada, Junichi, Morimoto, Masamichi
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US08/551,847
• Time in Patent Office: 1,316 Days
• Field of Search: 382/151, 382/181, 382/219, 382/220, 382/291, 382/209, 382/217, 382/180, 382/278, 382/199, 348/129, 348/130
• US Class Current: 382/151
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: G06V 10/757 Matching configurations of points or features

Abstract

When an image of an object is recognized, a template having a size smaller than the object is used. First, operation reference points C_s (X_m, Y_n) are set at steps smaller than lengths along the X and Y directions of an image of an object to be recognized, and a template T(x_i, y_i) is set at the operation reference points C_s (X_m, Y_n). Then, an equation, ##EQU1## is operated based on the template T(x_i, y_i) and the image data P(X, Y). Next, candidate points are determined as points having M(X_m, Y_n) larger than a reference value. Then, a position of the image of the object is determined according to the candidate points. Even if an image to be recognized is inclined, the position and the slope of the image can be detected at a fast speed even for image data with noise.

Claims

1. A method of pattern detection comprising the steps of:

obtaining an image of an object;

converting the image obtained into two-dimensional image data P(X, Y) at points (X, Y), wherein X and Y denote coordinates existing along two directions perpendicular to one another;

setting operation reference points C_s (X_m, Y_n) in two dimensions spaced from each other by a step size which is smaller than a size of the image of the object, wherein X_m and Y_n denote m-th and n-th coordinates existing along two directions and m and n denote integral numbers;

providing a two-dimensional template having a size smaller than the image of the object and having weights T(x_i, y_i) at a plurality of points (x_i, y_i) wherein x_i and y_i denote x and y coordinates of an i-th point and i denotes an integral number;

setting a position reference point of the template T(x_i, y_i) at each of the operation reference points C_s (X_m, Y_n);

calculating a weighted sum of the image data P(X, Y) by using the template T(x_i, y_i) for each of the operation reference points C_s (X_m, Y_n);

determining the operation reference points C_s (X_m, Y_n) as being candidate points when the weighted sum thereof is larger than a predetermined value;

grouping the candidate points into candidate groups comprising a plurality of adjacent candidate points, wherein the candidate groups sufficiently define a representative contour of the object; and

determining a position of the object based on positions of the candidate groups.

2. The method according to claim 1, wherein a size of the two-dimensional template is smaller than that of the image to be detected.

3. The method according to claim 1, wherein a shape and the size of the template T(x_i, y_i) are set according to a shape and pattern of the image of the object, a desired detection precision, and a desired detection speed.

4. The method according to claim 1, wherein the step size of the operation reference points C_s (X_m, Y_n) are set according to a shape and pattern of the image of the object to be detected, a desired detection precision, and a desired detection speed.

5. The method according to claim 1, wherein the template has weights which are non-zero.

6. The method according to claim 5, wherein the non-zero weights are equal.

7. The method according to claim 1, further comprising the step of detecting a center position of the candidate points in the candidate point group.

8. The method according to claim 7, wherein the object comprises a plurality of images to be detected, the method further comprising the step of detecting a slope of the images of the object based on the center positions of a plurality of the candidate point groups.