Self correcting portable digital radiography detector, methods and systems for same
Embodiments of radiographic imaging systems and/or methods can monitor the state of calibration of a digital x-ray detector, the detector including a solid state sensor with a plurality of pixels, an optional scintillating screen and at least one embedded microprocessor. In one embodiment, a method can use a computer or the embedded microprocessor or both, for setting a calibration operating mode of the portable detector; taking a plurality of dark images in the calibration mode; determining a dark difference image between pixel readings between two of the plurality of dark images; identifying pixels in the dark difference image that differ by over a threshold amount from at least some surrounding pixels in the dark difference image as defective pixels.
1. A method for monitoring the state of calibration of a digital x-ray detector, the detector comprising a solid state sensor with a plurality of pixels, a scintillating screen and at least one embedded microprocessor, the method comprising using a signal processing device, comprising:
setting a calibration operating mode of the detector including setting a voltage bias of the plurality of pixels at a first voltage level;
capturing a plurality of dark images in the calibration mode while maintaining the voltage bias of the plurality of pixels at the first voltage level;
resetting the voltage bias to a second voltage level different from the first voltage level only after the step of capturing the plurality of dark images;
determining a dark difference image between two captured dark images that were captured last in time of the captured plurality of dark images; and
identifying defective pixels in the dark difference image that differ by over at least one preselected threshold deviation amount from other pixels in the dark difference image.
2. The method according to claim 1 , where the two dark images have different integration times.
3. The method according to claim 1 , comprising:
turning on an imaging array of the detector; and
performing an initialization operation to inject charge into the pixels of the imaging array.
4. The method according to claim 1 , where identifying pixels comprises:
determining a local median at each pixel location in the dark difference image; and
determining a pixel is defective when a threshold measure between the local median at a pixel location and a pixel value at the pixel location in the dark difference image is exceeded.
5. The method according to claim 4 ,
wherein the threshold deviation amount is a fraction of the median image or where the threshold deviation amount is a fraction of the robust standard deviation of the image.
6. The method according to claim 4 , where the local median is determined using a median filter that is a set of 4 adjacent pixels, a set of 8 adjacent pixels, a subset of adjacent pixels, a filter of a prescribed size, a X-by-X square filter where X is an integer greater than 2, a X-by-Y rectangular filter where Y is an integer greater than 2 and not equal to X.
7. The method according to claim 1 , comprising initiating the calibration mode occurs after detected shock events to the detector, periodically, after a set number of exposures of the detector have been taken, after a time has elapsed, based on temperature, or by operator action.
8. The method according to claim 1 , comprising;
replacing a current updated defect map with the newly identified defective pixels to generate a replacement updated defect map; and
combining the replacement updated defect map with a factory defect map.
9. The method according to claim 1 , comprising correcting portions of radiographic images for any identified defective pixels.
10. The method according to claim 1 , comprising:
powering up the imaging array of the detector;
obtaining at least one of:
a first dark difference image using two dark images obtained with different integration times;
a second dark difference image using two or more dark images obtained by performing N initialization cycles before a first of the two or more dark images and by performing M initialization cycles before a second of the two or more dark images, where M and N are positive integers and the initialization cycles comprises at least one fill and one spill cycle;
a third dark difference as the second dark difference but with a different integration time;
a fourth dark difference image using at least two dark images obtained by performing an initialization operation to inject charge into the pixels;
a fifth dark difference image using more than one dark image obtained by performing a first refresh operation before a first of the more than one dark image and a second refresh operation before a second of the more than one dark image, where the second refresh operation is different from the first refresh operation;
powering down the detector between all sets of dark images; and
identifying defective pixels using one or more of the first through the fifth dark difference image.
11. The method according to claim 10 , where a pixel is identified as defective when a combination of two or more of the first dark difference image through the fifth dark difference image identify the pixel as defective.
12. The method according to claim 1 , where at least one of the two captured dark images comprises combined values or averaged values of multiple dark images obtained under the same conditions, further comprising correcting the two dark images for defect, where the signal processing device is a computer or an embedded microprocessor.
13. A system for monitoring the state of calibration of a digital x-ray detector, the detector comprising a solid state sensor with a plurality of pixels and at least one embedded microprocessor, the system comprising:
a computer operable during normal diagnostic use of the detector for accessing a digital image including digital image data of the detector;
the at least one embedded microprocessor configured to set a calibration operating mode of the detector including setting a voltage bias of the plurality of pixels at a first voltage level;
means for initializing the portable detector in the calibration mode and capturing a plurality of dark images in the calibration mode while maintaining the voltage bias of the plurality of pixels at the first voltage level;
means for resetting the voltage bias to a second voltage level different from the first voltage level only after capturing the plurality of dark images; and
the at least one embedded microprocessor or the computer configured to determine a dark difference image between two captured dark images that were captured last in time of the captured plurality of dark images and identify defective pixels in the dark difference image that differ by over a threshold amount from other pixels in the dark difference image.
14. The system according to claim 13 , where the detector is portable, comprising a scintillating screen, a battery and supporting a wireless link to the computer, where the means for initializing powers up the digital x-ray detector.
15. The system according to claim 13 , where the detector comprises:
at least one shock sensor for monitoring mechanical shock to the detector, and the system further comprises means for alerting a system user to conduct a calibration of the detector when a preselected shock threshold value has been exceeded.
16. A calibration system for monitoring the state of calibration of a radiographic detector, the detector comprising a solid state sensor with a plurality of pixels and at least one embedded microprocessor, the detector comprising:
a computer operable during normal diagnostic use of the detector for accessing a digital image including digital image data of the detector;
a defect calibration map comprising a single file including a plurality of bitplanes, the bitplanes comprising,
a first bitplane configured to store a factory calibration,
at least one second bitplane configured to store a second operator updated calibration that is repeatedly updated, and
at least one third bitplane configured to store a third calibration update that is repeatedly updated without the operator,
where the at least one third bitplane is updated more frequently than the at least one second bitplane, where the defect calibration map is accessible to the computer.
17. The calibration system of claim 16 , where the at least one second bitplane comprises one bitplane to store a flat field defect map update generated from a flat field calibration procedure and another bitplane to store a gain defect map update generated from a gain calibration procedure, and where the at least one third bitplane comprises one bitplane to store one defect map update generated from a first dark image procedure and another bitplane to store a defect map update generated from a second different dark image procedure, where a composite defect calibration map is generated from logically combining the first bitplane, the at least one second bitplane, and the at least one third bitplane, and where each bitplane comprises an ordered array.
18. A method of operating a digital radiographic detector comprising an array of imaging pixels, the method comprising:
powering up the array of imaging pixels;
obtaining a first dark difference image using two dark images each obtained using a different integration time, wherein the integration time is a time duration during which accumulated charges in the imaging pixels caused by a radiographic exposure are measured, and wherein the first dark difference image is determined using the difference between the two dark images;
powering down the detector after obtaining said first dark difference image; and
identifying defective pixels in the array of imaging pixels using said first dark difference image.
19. The method according to claim 18 , further comprising:
obtaining a second dark difference image using a pair of dark images obtained by performing N initialization cycles before obtaining a first of the pair of dark images and by performing M initialization cycles before obtaining a second of the pair of dark images, wherein M and N are positive integers, and wherein each of the initialization cycles comprise at least one fill and one spill cycle, wherein the second dark difference image is determined using the difference between the pair of dark images; and
obtaining a third dark difference image determined using the difference between the pair of dark images wherein the pair of dark images are each obtained using a different integration time,
wherein the step of powering down the detector is performed after obtaining each of the first, second, and third dark difference images, and
wherein the step of identifying defective pixels includes using at least two of the first, second, and third dark difference images.
20. The method according to claim 19 , further comprising determining row and column locations of all defective pixels in the array of imaging pixels relative to a zero position in the array.