Image processing system and method thereof in which three dimensional shape is reproduced from two dimensional image data

• US 5,548,667 A
• Filed: 09/07/1994
• Issued: 08/20/1996
• Est. Priority Date: 05/24/1991
• Status: Expired due to Term

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Tu, Peter
• Primary Examiner(s): COUSO, YON JUNG
• Application Number:
  US08/301,665
• Time in Patent Office: 713 Days
• Field of Search: 382/115, 382/203, 382/285, 382/286, 382/288, 382/291, 382/199, 348/94, 395/125
• US Class Current: 382/285
• CPC Class Codes: G06T 7/70 Determining position or orientation of objects or cameras camera calibration G06T7/80
• CPC Class Codes: G06V 10/754 involving a deformation of the sample pattern or of the reference pattern; Elastic matching

Abstract

The image processing system has a unit for photographing an object in two dimensions, a feature extraction unit for extracting features from the two-dimensional image data from the photographing means, and a three-dimensional shape reproduction unit. The feature extraction unit refers to feature points given to the object to extract the features. The three-dimensional shape reproduction unit expresses the object by a dynamic equation, applies force from the feature extraction coordinates to the dynamic model to cause the dynamic model to change shape and supplement depth data, and to thereby reproduce the three-dimensional shape of the object. To increase the speed of the processing, it is desirable to divide the image data of the object into portions with little changes in shape and perform the processing for reproducing the three-dimensional shape for each mode.

Claims

1. A method for detecting fingers of a human hand, comprising the steps of:

reading the hand and outputting image data of the hand in a form of two-dimensional pixel data;

detecting edges of the hand on the basis of said pixel data;

determining image lines as lines having the greatest number of edge pixels;

determining the orientation line of the hand as the line having the average slope and Y intercept of said image lines;

searching the pixel data for edge pixels wherein the search is statistically biased in a direction parallel to the orientation line; and

clustering edge pixels defining said finger tips;

wherein said steps of determining image lines and determining the orientation line include

transforming a rectilinear coordinate system of said detected edges into a polar coordinate system,

accumulating edge points in the parameter space, and

obtaining the highest accumulated values for each of said lines to determine said orientation line of the hand.

2. A method of detecting fingers of a human hand comprising the steps of:

reading the hand and outputting image data of the hand in a form of two-dimensional pixel data;

detecting edges of the hand on the basis of said pixel data;

determining image lines as lines having greatest number of edge pixels;

determining the orientation line of the hand as the line having the average slope and Y intercept of said image lines;

searching the pixel data for edge pixels wherein the search is statistically biased in a direction parallel to the orientation line; and

clustering edge pixels defining said finger tips;

wherein said clustering step includes

choosing arbitrary locations in a cluster,

assigning the termination pixel of each search to a cluster location of the basis of a least euclidean distance, and

determining new cluster locations by computing the centroid of all said pixels assigned to the cluster.

3. A method of extracting features of an object, comprising the steps of:

photographing the object and outputting image data of the object in a form of two-dimensional pixel data;

detecting edges of the object on the basis of said pixel data;

determining image lines as lines having greatest number of edge pixels;

determining the orientation line of the hand as the line having the average slope and Y intercept of said image lines;

searching the pixel data for edge pixels wherein the search is statistically biased in a direction parallel to the orientation line; and

clustering edge pixels defining said finger tips;

wherein said steps of determining image lines and the determining the orientation line include

transforming a rectilinear coordinate system of said detected edges into a polar coordinate system,

accumulating edge points in the parameter space, and

obtaining the highest accumulated values for each of said lines to determine said orientation line of the object.

4. A method of extracting features of an object, comprising the steps of:

photographing the object and outputting image data of the object in a form of two-dimensional pixel data;

detecting edges of the object on the basis of said pixel data;

determining image lines as lines having greatest number of edge pixels;

determining the orientation line of the hand as the line having the average slope and Y intercept of said image lines;

searching the pixel data for edge pixels wherein the search is statistically biased in a direction parallel to the orientation line; and

clustering edge pixels defining said finger tips;

wherein said clustering step includes

choosing arbitrary locations in a cluster,

assigning the termination pixel of each search to a cluster location of the basis of a least euclidean distance, and

determining new cluster locations by computing the centroid of all of said pixels assigned to the cluster.