Image analysis – Image enhancement or restoration – Image filter
Reexamination Certificate
1998-12-21
2002-08-06
Lee, Thomas D. (Department: 2724)
Image analysis
Image enhancement or restoration
Image filter
C382S236000
Reexamination Certificate
active
06430318
ABSTRACT:
Image processing method including spatial and temporal noise filtering steps, and medical imaging apparatus for carrying out this method.
FIELD OF THE INVENTION
The invention relates to an image processing method for reducing the noise in an image, including determination of three temporal intensities relating to a current pixel in the same location in three successive images in a sequence, the image to be processed being the central image of the sequence, which method includes spatial and temporal filtering steps. The invention also relates to a medical apparatus for carrying out this method.
The method is applied in order to reduce the noise in a sequence of images while preserving the small objects in motion. In this context small objects are to be understood to mean objects from 1 to a few pixels, for example from 1 to 10 pixels. The invention can be used particularly for the processing of video images, notably medical images.
BACKGROUND OF THE INVENTION
An image processing method for reducing the noise in a sequence of images which includes spatial and temporal filtering is already known from the publication “Multistage Order Statistic Filters for Image Sequence Processing” by Gonzalo R. ARCE, IEEE Transactions on SIGNAL PROCESSING, Vol.39, No. 5, MAY 91.
The cited document describes a median filtering method with several so-called MOS levels which combines the outputs of elementary spatial filters which operate in a cascade-type filtering structure comprising several levels. These elementary filters are conceived to match the structure extending in the window of the filter. In a spatial and temporal signal space each elementary filter is conceived to preserve a feature having substantially identical grey levels in a given direction (dimension). When a sufficiently large number of elementary filters is used, a feature oriented in any arbitrary direction can be preserved by the filter. The type of feature to be preserved determines the type of elementary filter of the MOS. If the feature is a unidirectional segment in a three-dimensional spatial and temporal space (cube), the filter MOS is called a unidirectional multi-level filter because of the fact that the elementary filter is unidirectional. If the feature has segments extending in two orthogonal directions, i.e. one in space and the other in time, the elementary filter is bi-directional and the resultant filter MOS is called a bi-directional multi-level filter. In an alternative version, a generalization of the class of filters described above is obtained by providing a degree of smoothing control by variation of the weight at the center of the window of the filter.
According to the cited document, these filters MOS are conceived to enable the restoration of details of the order of a pixel when the images of the sequence are free from motions causing an object to be displaced from one image to another. The filters MOS preserve the structure of the signal without taking into account motions of the object. The robustness with respect to noise, therefore, is dependent on the amplitude of the motions in the sequence.
SUMMARY OF THE INVENTION
It is an object of the invention to provide an image processing method for reducing the noise in a sequence of images representing small objects in motion while avoiding the formation of noise patches and patterns which are generally caused by spatial filtering of the current image in such an image sequence.
It is also an object of the invention to provide a method which avoids the dilemma between the elimination of noise patterns, which involves problems as regards preservation of objects in motion, and the preservation of objects in motion, involving problems in respect of residual noise patterns.
This object is achieved and the described problem is solved by means of an image processing method for reducing the noise in an image as claimed in claim
1
.
It is an advantage of this filtering method that it can be applied in circumstances in which the sequence of images contains very small objects which undergo very large motions. Generally speaking, this filtering method can operate in particularly difficult circumstances in respect of motion of the objects in the sequence of images. For example, this filtering method can be used in an image processing system which is included in a medical X-ray imaging apparatus operating in the fluoroscopy mode in which a sequence of images comprises only from 2 to 3 images per second, in which the images are extremely noisy because of the very low X-ray dose used for their formation, in which the movements of the objects from one image to another in the sequence are very large because of the comparatively long period of time elapsing between the formation of two successive images, and in which the objects in motion are also tools, such as catheters, which are extremely small. Even in these extremely difficult circumstances, very good results are obtained by means of this method. Thus, the method is capable of extracting the noise from a sequence of fluoroscopic images in the very difficult circumstances described above in such a manner that these noise-suppressed images correspond to images which would require an X-ray exposure time amounting to four times that of the images to be processed.
It is another advantage of the method that it does not require a priori knowledge of a variable which is called the standard noise deviation relative to the mean noise; such knowledge is necessary for carrying out many other known noise reduction methods.
It is another advantage of the method that it does not require any initial condition. It can be applied directly to a sequence of images.
A medical imaging apparatus which includes means for carrying out this method is defined in claim
10
.
REFERENCES:
patent: 4992995 (1991-02-01), Favret
patent: 5068799 (1991-11-01), Jarrett, Jr.
patent: 5793702 (1998-08-01), Paffenholz
patent: 5970180 (1999-10-01), Niihara et al.
patent: 01183980 (1989-07-01), None
Aufrichtig R et al: “X-Ray Fluoroscopy Spatio-Temoral Filtering With Object Detection”, IEEE Transactions on Medical Imaging, vol. 14, No. 4, Dec. 1, 1995, pp. 733-746.*
Brailean J C et al: “Simultaneous Recursive Displacement Estimation and Restoration of Noisy-Blurred Image Sequences”, IEEE Transactions on Image Processing, vol. 4, No. 9, Sep. 1, 1995, pp. 1236-1251.*
“Multistage Order Statistic Filters for Image Sequence Processing” by Gonzalo R. Arce, IEEE Transactions on Signal Processing, vol. 39, No. 5, May 1991.
Charpail François R. G.
Florent Raoul
Soyer Christel
Brinich Stephen
Charpail François R. G.
Koninklijke Philips Electronics , N.V.
Lee Thomas D.
Vodopia John
LandOfFree
Image processing method including spatial and temporal noise... 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 processing method including spatial and temporal noise..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing method including spatial and temporal noise... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2933021