Image analysis – Image enhancement or restoration – Artifact removal or suppression
Reexamination Certificate
1998-10-30
2002-04-30
Lee, Thomas D. (Department: 2624)
Image analysis
Image enhancement or restoration
Artifact removal or suppression
C382S271000
Reexamination Certificate
active
06381374
ABSTRACT:
TECHNICAL FIELD
The present invention relates to discrete pixel imaging systems, such as digital x-ray systems. More particularly, the invention relates to the identification of potentially defective circuitry in detection and signal processing components in such systems.
BACKGROUND ART
A number of techniques have been developed for producing images of an array of picture elements or pixels. Such systems typically include data acquisition circuitry which produces a series of signals representative of characteristics of pixels arranged in a matrix or array. For example, in a digital x-ray system, detector circuitry senses an amount of radiation impacting regions defining discrete pixels in a detector array. Based upon these radiation levels, the circuitry generates signals representative of each pixel, typically in the form of digitized intensity values. The values are transmitted to signal processing circuitry for filtering and enhancement. This circuitry may, for example, adjust the dynamic range of the detected values, and perform enhancement functions to emphasize or de-emphasize certain features of the image as defined by the pixel data. After processing, the data is stored for reconstruction of a composite image. In medical imaging systems, the composite image is useful to attending physicians and radiologists in diagnosis and treatment.
Detector circuitry in discrete pixel imaging systems is generally required to operate within specified tolerances. For example, in x-ray systems, circuitry associated with different pixels in the image matrix should produce output signals which are uniform for uniform levels of radiation received within the regions defined by the pixels. The uniformity provides accurate and repeatable image quality, permitting features in single images to be compared to one another, and images in a series to be similarly compared. However, some amount of deviation or tolerance is permitted between output signals for individual pixels to account for manufacturing differences, sensitivity differences, and so forth. Nevertheless, significant differences are generally not desirable between pixel circuitry output. Such differences can result in skewing of scaling of dynamic ranges, errors in image enhancement, production of anomalies or artifacts in the reconstructed image, and so forth.
Techniques have been developed for analyzing detector circuitry in discrete pixel imaging systems for potentially defective pixel circuits. In general, such techniques are implemented at final stages in the manufacture of the detectors or associated circuitry, or during calibration of the detector circuitry in the imaging system. Pixels identified as potentially defective may be flagged or masked in signal processing software. Thereafter, signals produced by the defective pixels may be disregarded or filtered separately from the non-defective pixel signals to reduce the likelihood of image processing errors or of unwanted artifacts.
While such techniques are useful in identifying potentially defective pixels, they are not always suitable for rapid reevaluation of discrete pixel detection systems after they are placed in service. Once a system is in service, it may be desirable periodically to perform a procedure for generating or updating a defective pixel mask or similar data used by the image processing circuitry. In particular, because detector circuitry can become defective after being placed in service, and thereby result in degradation of the image quality over time, it would be useful to provide a computationally efficient and straightforward technique for identifying potentially defective pixel circuits that could be implemented by operations personnel, clinicians or technicians. Pixels newly identified as potentially defective could then be flagged or added to a mask employed by image processing software. If an unacceptable degree of degradation is detected through the technique, the detector or associated circuitry may be replaced to once again provide the desired image quality.
DISCLOSURE OF THE INVENTION
The invention provides a technique for detecting potentially defective pixel circuits designed to respond to these needs. The technique may be employed in the evaluation of new detector circuitry, but is particularly well suited to analyzing existing circuitry already placed in service. Moreover, while the technique is particularly suitable for use in digital x-ray systems, it may find application in other imaging modalities wherein discrete pixel data are collected via detector circuitry wherein the output may vary from pixel-to-pixel within defined tolerances. The technique may be implemented in a straightforward procedure which is relatively time-efficient. Processing steps used to implement the technique may be programmed into existing imaging system controllers, further facilitating its use in existing systems. Pixels having output levels which differ in statistically significant ways from other pixels in the system are flagged and may be added to a map or mask used by image processing circuitry.
REFERENCES:
patent: 5321497 (1994-06-01), Ai et al.
patent: 5537669 (1996-07-01), Evans et al.
patent: 5617461 (1997-04-01), Schreiner
patent: 5877501 (1999-03-01), Ivan et al.
patent: 5917935 (1999-06-01), Hawthorne et al.
Brinich Stephen
Fletcher Yoder & Van Someren
Lee Thomas D.
LandOfFree
Histogram analysis method for defective pixel identification does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Histogram analysis method for defective pixel identification, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Histogram analysis method for defective pixel identification will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2917697