Differences

This shows you the differences between two versions of the page.

--- uvscada:gxs700 [2017/12/23 18:40] – [Table] mcmaster
+++ uvscada:gxs700 [2017/12/31 19:19] – [Calibration] mcmaster
@@ Line 170: / Line 170: @@
 Takes raw, uncalibrated images.  I've been asked a few times if I could provide calibration files for the Windows driver or could even use their calibration files on my driver.  However, I'm not really interested in supporting their proprietary format. If someone submits a patch though I'll probably merge it in.
-That said, I'm not against coming up with an independent calibration scheme.  I know how to take dark frames (see API) which is the key component.  Uncalibrated images have been generally good enough for me so I haven't put work into this.  If you are interested though shoot me a line.
+===== cal.py flow =====
+Quick scheme to get rid of the worst artifacts
+Basic idea:
+  - Capture a dark field frame (no x-ray) to get the lowest possible pixel value
+  - Capture a flat field frame (x-ray with no object) to get the highest possible pixel value
+  - Rescale images to range found above
+Notes:
+  * Decoded images invert pixel values (per x-ray industry convention), so high and low are relative
+  * As of late 2017 capture.py now spits out a sensor .json file with each image. Hopefully the scanner will add this soon as well
+  * As of this writing no bad pixel replacement is done
+==== 1: Capture dark field ====
+First, take a dark field capture using something like:
+  $ gxs700-capture -f -e
+This should yield a raw picture that looks something like this:
+{{:uvscada:gxs700:cal:df.jpg?300|}}
+And histogram equalized looks like this:
+{{:uvscada:gxs700:cal:dfe.jpg?300|}}
+==== 2: Capture flat field ====
+Now make sure the sensor and the x-ray head have the exact same position and settings you will use to take your actual image capture. Make sure there is nothing in the beam and do a standard capture (and also firing the x-ray head):
+  $ gxs700-capture -e
+This produces a flat field image like this:
+{{:uvscada:gxs700:cal:ff.jpg?300|}}
+Which might look like this when histogram equalized:
+{{:uvscada:gxs700:cal:ffe.jpg?300|}}
+For best results you should also let the sensor and x-ray head sit a while to warm up. I'm not sure how long matters, but maybe 10 minutes would be good
+==== 3: Capture data ====
+ie using capture.py or xray_plan_cli.py
+Put your sample in and use the exact setup as the flat field to capture data. Don't move the sensor or x-ray head
+==== 4: Correct images ====
+run something like
+  $ python cal.py  df.png ff.png png cal
+Where:
+  * df.png is your dark field capture
+  * ff.png is your flat field capture
+  * png is a directory with your images to be corrected
+  * cal is the output directory with corrected images
+Here is a raw image as captured:
+{{:uvscada:gxs700:cal:ex_raw.jpg?300|}}
+And with histogram equalization:
+{{:uvscada:gxs700:cal:ex_raw_e.jpg?300|}}
+Now the corrected image:
+{{:uvscada:gxs700:cal:ex_out.jpg?300|}}
+And it with histogram equalization:
+{{:uvscada:gxs700:cal:ex_out_e.jpg?300|}}
 ====== Quality ======