Image analysis – Pattern recognition – Feature extraction
Reexamination Certificate
1998-12-11
2001-06-26
Patel, Jay (Department: 2721)
Image analysis
Pattern recognition
Feature extraction
C382S164000, C382S283000
Reexamination Certificate
active
06252985
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to an image area extracting method and is suitably applied, for example, to the case of forming a key signal (a mask image) which is necessary at the time of an image synthesizing process among the works related to video production.
2. Description of the Related Art
Techniques to extract an object from an image are conventionally well-known as methods for special effects. The techniques to extract an object are the mask image hand painting (video mat) method, the method using chroma key processing and luminance key processing, and the rough contour assigning and thinning (Object Extracting Method for Image Mixing, by Seiki Inoue, The transactions of the Institute of Electronics, Information, and Communication Engineers D-II vol.j74-D-II pages 1411-1418, 1991).
The mask image hand painting method, in which an operator assigns an object area by hand painting in utilizing a painting tool such as a tablet, is a method capable of inputting the boundary of an object precisely.
On the other hand, the chroma key processing or the luminance key processing is the method to divide the area with a specified color or specified luminance by performing threshold processing on color information or luminance information of the input image. In this method, only assigning of the specified color or the specified luminance by the operator makes it possible to execute a uniform processing automatically to all picture elements.
Moreover, in the method of rough contour assigning and thining, the rough area including the contour of the object is assigned by an operator to obtain the shape of object by thining the rough area based on the intensity of edge so as to match this rough area to the contour of object automatically.
However, in the chroma key processing and the luminance key processing, an input image of which its characteristics greatly differ according to the position in one screen is processed in accordance with single threshold value to the pixel value of the entire image, the chroma key processing and the luminance key processing cannot be applied to the image with the arbitrary background. Therefore, they can be applied only to the specified input image taken with the uniform background for the purpose of synthesizing.
The rough contour assigning and thinning method is the method to assign the rough area and which is capable of extracting the object area from the uneven background. However, since this method is the processing only dependent upon the edge intensity, the correct result cannot be obtained in the area with an intense edge which is not the contour of the object, and there are cases where the accuracy is worse as compared with the video mat method. Also, in the case where the characteristics of background differs according to the position on the image, since single area extracting method is used, it is impossible to cope with the arbitrary background thoroughly.
Accordingly in these methods, it was only video mat method which was capable of extracting the object area accurately from the optional background. However, because the video mat method requires the process to assign the object area per one pixel, it requires much time and skill and there arises a problem that its operational efficiency is not favorable.
OBJECTS AND SUMMARY OF THE INVENTION
In view of the foregoing, an object of this invention is to provide an image area extracting method capable of extracting the desired area from the image accurately and efficiently.
The foregoing object and other objects of the present invention have been achieved by the provision of an image area extracting method for extracting the desired area from the image, comprising the steps of: dividing the rough boundary area corresponding to the image into a plurality of small areas, in which the boundary area between the desired area and the area other than the desired area is roughly assigned; executing the area extracting procedure to each small area to detect the desired area and areas other than the desired area from each small area; and forming the boundary area mask image based on the detection result and extracting the desired area based on the boundary area mask image.
The rough boundary area image in which the boundary area between the desired area and areas other than the desired area is roughly assigned is divided into a plurality of small areas, then the area extracting procedure is executed to each small area to detect the desired area and the areas other than the desired area from each small area, then the boundary area mask image is formed based on the result of above detection, and the desired area is extracted based on the boundary area mask image. Thus, the desired area can be extracted accurately and efficiently from the image.
The nature, principle and utility of the invention will become more apparent from the following detailed description when read in conjunction with the accompanying drawings in which like parts are designated by like reference numerals or characters.
REFERENCES:
patent: 4814755 (1989-03-01), Johnson
patent: 4868884 (1989-09-01), Miyazaki
patent: 5253339 (1993-10-01), Wells et al.
patent: 5386234 (1995-01-01), Veltman et al.
patent: 5440652 (1995-08-01), Ting
patent: 5461420 (1995-10-01), Yonemitsu et al.
patent: 5481553 (1996-01-01), Suzuki et al.
patent: 5515388 (1996-05-01), Yagasaki
patent: 5574800 (1996-11-01), Inoue et al.
patent: 5596655 (1997-01-01), Lopwz
patent: 5604822 (1997-02-01), Pearson et al.
patent: 5617487 (1997-04-01), Yoneyama et al.
patent: 5630037 (1997-05-01), Schindler
patent: 5687251 (1997-11-01), Erler et al.
patent: 5892853 (1999-04-01), Hirani et al.
Benson, Television Engineering Handbook, McGraw Hill, New York, Published Dec. 1985, pp. 2.7 & 2.8.
Beyerer, J. “Suppression of Stochastically Placed, Straignt Toolmarks To Enhance Objects and Defects”, Techniques Messen TM 1982-1988, vol. 59, No. 10, Oct. 1, 1992, pp. 389-397.
Ferreira, P., et al., “Errorless Restoration Algorithms For Band-Limited Images”, IEEE Nov. 1994, pp. 157-161.
Fienup, J.R., “Phase Retrieval; Algorithms: A Comparison”, Applied Optics, vol. 21, No. 15, pp. 2758-2769, Aug. 1, 1982.
Gerchberg, R.W., “Super-Resolution Through Error Energy Reduction”, Optica Acta, 1974, vol. 21, No. 9, 709-720, (Received Sep. 13, 1973, Revision Received Jan. 15, 1974).
Inoue, S., “An Object Extraction Method For Image Synthesis”, Proceedings of the SPIE, The International Society for Optical Engineering, Conference Date Nov. 1991, vol. J74-D-11, No. 10, pp. 1411-1418.
Jain, A., “Extrapolation Algorithms For Discrete Signals With Application In Spectral Estimation”, IEEE Transactions On Acoustics, Speech, And Signal Processing, vol. ASSP-29, No. 4, pp. 830-845, Aug. 1981.
Pang, J., et al., “Signal Recovery From Subband Energy Constraints”, Optical Society Of America, vol. 9, No. 7, pp. 1041-1051, Julk. 1992.
Papoulis, A., “A New Algorithm In Spectral Analysis And Band-Limited Extrapolation”, IEEE Transactions On Circuits And Systems, vol. CAS-22, No. 9, pp. 735-742, Sep. 1975.
Peng, H., et al., “Signal Recovery with Similarity Constraints”, Optical Society of America, vol. 6, No. 6, pp. 844-851, Jun. 1989.
Popat, K., et al., “Novel Cluster-Based Probability Model for Texture Synthesis, Classification, and Compression,” Massachusetts Institute of Technology, Media Laboratory, SPIE vol. 2094, pp. 756-768, Nov. 1993.
Schafer R., et al., “Constrained Iterative Restoration Algorithms”, Proceedings Of The IEEE, vol. 69, No. 4, pp. 432-450, Apr. 1981.
Sezan, M., et al., “Image Restoration By The Method Of Convex Projections: Part 2-Applications and Numerical Results”, IEEE Transactions On Medical Imaging, vol. MI-1, No. 2k, pp. 95-101, Oct. 1982.
Sezan, M., et al., “Prototype Image Constraints For Set-Theoretic Image Restoration”, IEEE Transactions On Signal Processing, vol. 39, No. 10, pp. 2275-2285, Oct. 1991.
Youla,. D. C., et al., “Image Restoration By the Method of Convex Projections: Part 1-Theory, IEEE Transactions on Medical Imaging”,
Mitsunaga Tomoo
Totsuka Takashi
Bell Boyd & Lloyd LLC
Patel Jay
Sony Corporation
LandOfFree
Image detecting apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image detecting apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image detecting apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2518494