Apparatus and method for image processing

• US 6,996,255 B2
• Filed: 05/24/2004
• Issued: 02/07/2006
• Est. Priority Date: 05/28/1999
• Status: Expired due to Fees

• Current Assignee: Nippon Telegraph and Telephone Corporation
• Original Assignee: Nippon Telegraph and Telephone Corporation
• Inventors: Sakuma, Satoshi, Takahashi, Yasuko, Ohtsuka, Sakuichi, Shio, Akio
• Primary Examiner(s): Couso, Jose L.
• Assistant Examiner(s): Lu, Tom Y.
• Application Number:
  US10/852,873
• Publication Number:
  US 20050008194A1
• Time in Patent Office: 624 Days
• Field of Search: 382103-107, 340933-937, 348/149, 702/142
• US Class Current: 382/107
• CPC Class Codes: G01P 3/806 in devices of the type to be classified in G01P3/68
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G08G 1/04 using optical or ultrasonic detectors

Abstract

The method for measuring speed is performed by dividing the time sequence image data A into long and narrow frames (step 401). Next, the framed images and the time sequence image data B produced by the line scan camera B are correlated to obtain a moving time (step 402). Next, a moving time of each framed image and a corresponding moving distance are obtained to compute a moving speed (step 403). Next, scale of a specific shape pattern and the like is corrected using the moving speed obtained to match the scales in the two image sequences A and B to obtain a similarity degree (step 404). Next, a threshold value is applied to the similarity degree to recognize the specific shape pattern, thereby recognizing that the images are produced by a moving vehicle.

Claims

1. A method for processing image data of an object body containing a specific shape pattern that is recognizable as an image and moving along a given track relative to an observation point, the method comprising the steps of:

obtaining time sequence image data recorded by synchronously operating line image acquiring apparatuses arranged transversely to the given track to record time sequence images appearing between parallel line axes at a constant timing determined by a line scanning cycle;

computing a size of the line displacement of the object body on the time axis between image acquiring positions of the line image acquiring apparatuses, by correlating the object body between the framed images of time sequence image data, and obtaining a moving time interval according to the computed value of the line displacement and the line scanning cycle;

obtaining a relative speed of the object body according to the moving time interval and a distance separating the image acquiring positions of the line image acquiring apparatuses arranged on the given track;

matching a time scale of the time sequence image data and a time scale of a template for the specific shape pattern by correcting respective time scales by dividing the time axis in each time sequence image data by the relative speed;

computing a similarity degree between the time sequence image data and the template selected from predefined specific shape patterns so as to detect the specific shape pattern contained in the object body in the time sequence image data and identify the object body; and

obtaining the relative speed of the object body by recalculating the moving time interval of the portion of the specific shape pattern.

2. A method for processing image data of an object body containing a specific shape pattern that is recognizable as an image and moving along a given track relative to an observation point, the method comprising the steps of:

obtaining time sequence image data recorded by synchronously operating a plurality of line image acquiring apparatuses, arranged transversely to the given track of the moving object body, so as to record time sequence images appearing between two parallel line axes at a constant timing determined by a line scanning cycle;

correlating framed images of the object body in the time sequence image data by computing a similarity degree, and obtaining a moving time interval of the object body, to move between the line image acquiring apparatuses, from the correlated images according to the value of line displacement of the object body and the line scanning cycle;

obtaining a relative speed of the object body relative to the observation point according to the moving time interval and a distance separating image acquiring positions of the line image acquiring apparatuses on the given track;

matching a time scale of the time sequence image data and a time scale of a template for the specific shape pattern by correcting respective time scales by dividing the time axis in each time sequence image data by the relative speed;

computing a similarity degree between the time sequence image data and the template selected from predefined specific shape patterns so as to detect the specific shape pattern contained in the object body in the time sequence image data and identify the object body; and

obtaining the relative speed of the object body by recalculating the moving time interval of the portion of the specific shape pattern.

3. A recording medium capable of being read by a computer, upon which is recorded a computer program for causing the computer to execute the method steps of:

obtaining time sequence image data recorded by synchronously operating line image acquiring apparatuses arranged transversely to the given track to record time sequence images appearing between parallel line axes at a constant timing determined by a line scanning cycle;

computing a size of the line displacement of the object body on the time axis between image acquiring positions of the line image acquiring apparatuses, by correlating the object body between the framed images of time sequence image data, and obtaining a moving time interval according to the computed value of the line displacement and the line scanning cycle;

obtaining a relative speed of the object body according to the moving time interval and a distance separating the image acquiring positions of the line image acquiring apparatuses arranged on the given track;

matching a time scale of the time sequence image data and a time scale of a template for the specific shape pattern by correcting respective time scales by dividing the time axis in each time sequence image data by the relative speed;

computing a similarity degree between the time sequence image data and the template selected from predefined specific shape patterns so as to detect the specific shape pattern contained in the object body in the time sequence image data and identify the object body; and

obtaining the relative speed of the object body by recalculating the moving time interval of the portion of the specific shape pattern.

4. A recording medium capable of being read by a computer, upon which is recorded a computer program for causing the computer to execute the method steps of:

obtaining time sequence image data recorded by synchronously operating a plurality of line image acquiring apparatuses, arranged transversely to the given track of the moving object body, so as to record time sequence images appearing between two parallel line axes at a constant timing determined by a line scanning cycle;

correlating framed images of the object body in the time sequence image data by computing a similarity degree, and obtaining a moving time interval of the object body, to move between the line image acquiring apparatuses, from the correlated images according to the value of line displacement of the object body and the line scanning cycle;

obtaining a relative speed of the object body relative to the observation point according to the moving time interval and a distance separating image acquiring positions of the line image acquiring apparatuses on the given track;

matching a time scale of the time sequence image data and a time scale of a template for the specific shape pattern by correcting respective time scales by dividing the time axis in each time sequence image data by the relative speed;

computing a similarity degree between the time sequence image data and the template selected from predefined specific shape patterns so as to detect the specific shape pattern contained in the object body in the time sequence image data and identify the object body; and

obtaining the relative speed of the object body by recalculating the moving time interval of the portion of the specific shape pattern.