Image processing apparatus and method

An image processing apparatus, comprising:

• moving-image input means for inputting a moving image composed of a plurality of frames;
• background-image acquisition means for acquiring a background image relating to the moving image that has been input;
• object extraction means for extracting an object by comparing each of the frames constituting the moving image with the background image;
• abnormal-data discrimination means for discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• shape-data correction means for correcting the shape data based upon a result of discrimination of the shape data;
• image-data correction means for generating image data, which represents the image of the object, conforming to the shape data that has been corrected by said shape-data correction means;
• and
• coding means for coding the shape data and the image data,
• wherein said abnormal-data discrimination means discriminates the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination means comprises:
• first comparison means for comparing the shape data of the object using a present frame and a preceding frame;
• second comparison means for comparing the shape data of the object using the present frame and a succeeding frame;
• and
• decision means for deciding that correction of the present frame is necessary if a difference in the shape data of the object between the present frame and the preceding frame is large and a difference in the shape data of the object between the present frame and the succeeding frame is large.

An image processing apparatus, comprising:

• moving-image input means for inputting a moving image composed of a plurality of frames;
• background-image acquisition means for acquiring a background image relating to the moving image that has been input;
• object extraction means for extracting an object by comparing each of the frames constituting the moving image with the background image;
• abnormal-data discrimination means for discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• shape-data correction means for correcting the shape data based upon a result of discrimination of the shape data;
• image-data correction means for generating image data, which represents the image of the object, conforming to the shape data that has been corrected by said shape-data correction means;
• and
• coding means for coding the shape data and the image data,
• wherein said abnormal-data discrimination means discriminates the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination means comprises:
• first comparison means for comparing the shape data of the object using a present frame and a preceding frame;
• second comparison means for comparing the shape data of the object using frames following the present frame and the preceding frame;
• and
• decision means for deciding that correction of prescribed frames from the present frame onward is necessary if a difference between the present frame and the preceding frame is large and a difference in the shape data of the object between the preceding frame and the frames following the present frame is small.

An image processing apparatus, comprising:

• moving-image input means for inputting a moving image composed of a plurality of frames;
• background-image acquisition means for acquiring a background image relating to the moving image that has been input;
• object extraction means for extracting an object by comparing each of the frames constituting the moving image with the background image;
• abnormal-data discrimination means for discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• shape-data correction means for correcting the shape data based upon a result of discrimination of the shape data;
• image-data correction means for generating image data, which represents the image of the object, conforming to the shape data that has been corrected by said shape-data correction means;
• and
• coding means for coding the shape data and the image data,
• wherein said shape-data correction means corrects the shape data of the object in an Nth frame, which has been determined to be abnormal, using an (N−1)th or earlier frame determined to be normal and an (N+1)th or later frame determined to be normal, and
• wherein said shape-data correction means comprises detection means for detecting corresponding points in the shape data between an (N−1)th or earlier frame and an (N+1)th or later frame.

An image processing method, comprising:

• a moving-image acquisition step of acquiring a moving image composed of a plurality of frames;
• a background-image acquisition step of acquiring a background image relating to the moving image;
• an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in said shape-data correction step;
• and
• a coding step of coding the shape data and the image data,
• wherein said abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using the preceding frame and a succeeding frame;
• and
• a decision step of deciding that correction of the present frame is necessary if a difference in the shape data of the object between the present frame and the preceding frame is large and a difference in the shape data of the object between the preceding frame and the succeeding frame is small.

An image processing method, comprising:

• a moving-image acquisition step of acquiring a moving image composed of a plurality of frames;
• a background-image acquisition step of acquiring a background image relating to the moving image;
• an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in said shape-data correction step;
• and
• a coding step of coding the shape data and the image data,
• wherein said abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using the present frame and a succeeding frame;
• and
• a decision step of deciding that correction of the present frame is necessary if a difference in the shape data of the object between the present frame and the preceding frame is large and a difference in the shape data of the object between the present frame and the succeeding frame is large.

An image processing method, comprising:

• a moving-image acquisition step of acquiring a moving image composed of a plurality of frames;
• a background-image acquisition step of acquiring a background image relating to the moving image;
• an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in said shape-data correction step;
• and
• a coding step of coding the shape data and the image data,
• wherein said abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using frames following the present frame and the preceding frame;
• and
• a decision step of deciding that correction of prescribed frames from the present frame onward is necessary in a case where a difference between the present frame and the preceding frame is large and a difference in the shape data of the object between the preceding frame and the frames following the present frame is small.

An image processing method, comprising:

• a moving-image acquisition step of acquiring a moving image composed of a plurality of frames;
• a background-image acquisition step of acquiring a background image relating to the moving image;
• an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• a shape-data correction step of correcting the shape data based upon result of discrimination of the shape data;
• an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in said shape-data correction step;
• and
• a coding step of coding the shape data and the image data,
• wherein said shape-data correction step includes correcting the shape data of the object in an Nth frame, which has been determined to be abnormal, using an (N−1)th or earlier frame determined to be normal and an (N+1)th or later frame determined to be normal, and
• wherein said shape-data correction step further includes a detection step of detecting corresponding points in the shape data between an (N−1)th or earlier frame and an (N+1)th or later frame.

A computer-readable storage medium storing a program which, when executed, performs a method for processing an image, the program comprising:

• code for a moving-image input step of inputting a moving image composed of a plurality of frames;
• code for a background-image acquisition step of acquiring a background image relating to the moving image that has been input;
• code for an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• code for an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• code for a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• code for an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in the shape-data correction step;
• and
• code for a coding step of coding the shape data and the image data,
• wherein said abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein said abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using the preceding frame and a succeeding frame;
• and
• a decision step of deciding that correction of the present frame is necessary if a difference in the shape data of the object between the present frame and the preceding frame is large and a difference in the shape data of the object between the preceding frame and the succeeding frame is small.

A computer-readable storage medium storing a program which, when executed, performs a method for processing an image, the program comprising:

• code for a moving-image input step of inputting a moving image composed of a plurality of frames;
• code for a background-image acquisition step of acquiring a background image relating to the moving image that has been input;
• code for an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• code for an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• code for a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• code for an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in the shape-data correction step;
• and
• code for a coding step of coding the shape data and the image data,
• wherein the abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein the abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using the present frame and a succeeding frame;
• and
• a decision step of deciding that correction of the present frame is necessary if a difference in the shape data of the object between the present frame and the preceding frame is large and a difference in the shape data of the object between the present frame and the succeeding frame is large.

A computer-readable storage medium storing a program which, when executed, performs a method for processing an image, the program comprising:

• code for a moving-image input step of inputting a moving image composed of a plurality of frames;
• code for a background-image acquisition step of acquiring a background image relating to the moving image that has been input;
• code for an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• code for an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• code for a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• code for an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in the shape-data correction step;
• and
• code for a coding step of coding the shape data and the image data,
• wherein the abnormal-data discrimination step includes discriminating the necessity of correcting the shape data by comparing the shape data of the object from frame to frame of a plurality of frames that differ in time, and
• wherein the abnormal-data discrimination step further includes:
• a first comparison step of comparing the shape data of the object using a present frame and a preceding frame;
• a second comparison step of comparing the shape data of the object using frames following the present frame and the preceding frame;
• and
• a decision step of deciding that correction of prescribed frames from the present frame onward is necessary if a difference between the present frame and the preceding frame is large and a difference in the shape data of the object between the preceding frame and the frames following the present frame is small.

A computer-readable storage medium storing a program which, when executed, performs a method for processing an image, the program comprising:

• code for a moving-image input step of inputting a moving image composed of a plurality of frames;
• code for a background-image acquisition step of acquiring a background image relating to the moving image that has been input;
• code for an object extraction step of extracting an object by comparing each of the frames constituting the moving image with the background image;
• code for an abnormal-data discrimination step of discriminating whether shape data representing the shape of the extracted object is abnormal or not;
• code for a shape-data correction step of correcting the shape data based upon a result of discrimination of the shape data;
• code for an image-data correction step of generating image data, which represents the image of the object, conforming to the shape data that has been corrected in the shape-data correction step;
• and
• code for a coding step of coding the shape data and the image data,
• wherein the shape-data correction step includes correcting the shape data of the object in an Nth frame, which has been determined to be abnormal, using an (N−1)th or earlier frame determined to be normal and an (N+1)th or later frame determined to be normal, and
• wherein the shape-data correction means further includes a detection step of detecting corresponding points in the shape data between an (N−1)th or earlier frame and an (N+1)th or later frame.