Method for correcting defective pixel artifacts in a direct radiography image
A method for reducing image disturbances caused by reconstructed defective pixel clusters located in signal-gradient affected diagnostic image regions. An individually adapted central symmetrical pair reconstruction (CSP) kernel is composed for a defective image pixel based on a kernel-pair candidate order encoded in a model thereby using the pixel's validity state. The image impacted by defective pixels is corrected in real-time by statistical filtering or spatial convolution of the kernel-associated image data accessible via a predetermined CSP kernels image-offsets structure.
1. A method for correcting defective pixel artifacts in a direct radiography image comprising executing, for a defective pixel, the steps of:
gathering information on a defective state of image pixels surrounding the defective pixel;
composing an adaptive reconstruction kernel by combining a set of pair-wise arranged kernel locations which correspond to non-defective pixels in a vicinity of the defective pixel;
obtaining image data associated with the set of pair-wise arranged kernel locations in the adaptive reconstruction kernel;
processing the associated image data along with kernel data using a reconstruction algorithm to calculate a replacement value for image data of the defective pixel; and
replacing the defective pixel image data with the replacement value; wherein the adaptive reconstruction kernel includes an even number of non-defective, center symmetrical kernel-pair locations relative to a center of the adaptive reconstruction kernel and is added to a center symmetrical pair kernels structure.
2. The method according to claim 1 , wherein the reconstruction algorithm performs a statistical filtering process on the image data spatially associated with the set of pair-wise arranged kernel locations in the adaptive reconstruction kernel.
3. The method according to claim 1 , wherein the reconstruction algorithm performs a spatial convolution process on the image data spatially associated with the set of pair-wise arranged kernel locations in the adaptive reconstruction kernel using associated kernel weights.
4. The method according to claim 3 , wherein the spatial convolution process includes calculating a weighted average value of the image data associated with the adaptive reconstruction kernel using corresponding kernel weights.
5. The method according to claim 4 , wherein weights of the non-defective, center symmetrical kernel-pair locations relative to the center of the adaptive reconstruction kernel in the adaptive reconstruction kernel are determined based on a relative distance model wherein a kernel location weight decreases with an increasing distance between the kernel location and the center of the adaptive reconstruction kernel.
6. The method according to claim 1 , wherein the center symmetrical kernel-pair locations are composed by order-consecutively selecting a predetermined number of kernel-pair candidates from a center symmetrical pair model based on the defective state of kernel-pair associated sensor pixels.
7. The method according to claim 6 , wherein the center symmetrical pair model is arranged as an order-sorted set of the kernel pair candidates geometrically defined by eccentricities and angular directions relative to the center of the adaptive reconstruction kernel.
8. The method according to claim 7 , wherein the center symmetrical pair eccentricity is defined such that a distance from the center of the adaptive reconstruction kernel does not decrease with sorting order.
9. The method according to claim 7 , wherein the center symmetrical pair angular directions are mutually perpendicular per set of two order-sorted kernel pairs candidates.
10. The method according to claim 1 , wherein information on the defective state of the image pixels surrounding the defective pixel is derived from a defective pixel locations structure and is merged with image position information derived from the center symmetrical pair kernels structure to create an image-offsets structure that is used during the reconstruction.
11. A non-transitory computer readable medium comprising computer executable program code adapted to carry out, when the computer executable program code is executed on a computer, the steps of claim 1 .