HW7: Image Reduction Report

Image Reduction Report

Joe Regan

10/23/15

Abstract/Introduction:

On Wednesday, October 21, 2015, we used previously written code to reduce images we had taken at Baker Observatory a few weeks prior on September 28, 2015. Each student took images of a different object of their choosing, myself choosing the star Altair. In my final reduced images, the counts for the star reached up to roughly 24,000 counts at the brightest point. Due to imperfections in the image reduction process, we were left with some negative final pixel values in the reduced images, which were later corrected. 

Procedure:

This experiment's purpose was to reduce the images taken at Baker Observatory, and to familiarize students with the 8-inch CPC telescopes along with the reduction process. First, ten darks and flats were taken each at ten second exposure times, in order to detect the dark current, caused by electrical current, in the detector, and to evenly distribute the light in the final image, respectively. Next, ten science darks were taken as an extra precaution. Finally, in my case, ten science images were taken of the star Altair, each at 0.5 second exposures due to the brightness of the star.

Fig. 1: Example raw image. Highest value: 29516, Lowest value: 970

The following examples of darks, science darks, and flats were taken to subtract the dark current and bad pixels from the image.

Fig. 2: Example Dark image. Highest value: 20657, Lowest value: 891

Fig. 3: Example Science Dark. Highest value: 13620, Lowest value: 966

Figure 4: Example Flat image. Highest value: 33041, Lowest value: 13230

Results and Discussion:

The program we used was, in this case, reducestinb.pro, since there were no bias images taken at Baker. The program was written to subtract the bad pixels caused by the dark current, and to adjust pixel counts using the flat images. The program is not perfect, and was giving negative pixel values as low as roughly -24000 counts. Adjusting the program to account for these negative values, and adding 200 counts to the overall image to adjust for a negative average outside of the target area, the final reduced image was produced.

Fig. 5: Example reduced image. Highest value: 28614, Lowest value: -299.899

A notable difference is shown from the raw image, as the reduced image shows far less variation in brightness of the overall image, and the target stands out a bit more visibly due to the scaling of the image. The brightness of the target hardly changed, only slightly darker in the final reduced image. 

The negative counts are still there, possibly due to either an error in the reduction process or unexpected initial subtraction of the dark current, automatically produced by the detector. Both issues can be corrected with a slight change in the reduction code.

Reduced Image- Lab Report

AST 311: Observational Techniques

Reducing Astronomical Images

Missouri State University

Joshua W. Kern

Abstract

We reduced the science images using IDL and, having an exposure time of 0.05 seconds for our science images, we obtained counts of ~14,000 for Saturn. However, negative counts were given in all other areas of our reduced science image not associated with Saturn itself. This is a manifestation of the fact that while obtaining our science images an automatic dark correction was made on the image. Since our code subtracted the darks as well we, in effect, subtracted two sets of dark images. This will be corrected for in future imaging endeavors. Also, due to tracking errors with the telescope, Saturn drifted in our images.  This made creating a final stacked science image impossible without correcting for the drift. Therefore our final product is 10 separately reduced science images.

Experiment

The purpose of this experiment was to reduce the astronomical images we obtained on September 30, 2015 of Saturn at Baker Observatory. We used the 8-inch telescopes with the ST-I CCD cameras to obtain science images as well as all of the necessary correction images. An example of our raw science image is given below with counts of ~16,000 corresponding to Saturn itself.

Examples of our darks, biases, and flats used for image reduction are given below respectively.

Results and Discussion

To reduce our science images we used IDL to subtract off the excess counts caused by “noise” from the sky and the instruments. We also adjusted the counts for pixel to pixel variations by taking flats. Since no filter was used to obtain the science images, no filter was used when obtaining flats. Exposure times for darks and flats were both 2 seconds, and exposure time for our sky-dark and science images were 0.05 seconds.

We then used a code written by Dr. Peter Plavchan that utilizes this information to reduce our raw science image. An example of the reduced science image is given below with counts of ~14,000 corresponding to Saturn. These counts are slightly lower than the raw science image which is to be expected.

However, in areas of the image which don’t correspond to Saturn the counts have negative values. This is due to the fact that an automatic dark correction was made while obtaining the raw science images without our knowledge. Therefore, when we reduced our images using Dr. Plavchan’s code we consequently subtracted off the dark images a second time leaving negative values where the counts were already much smaller than the counts associated with Saturn. This can be easily corrected for future image reduction by slightly modifying the image reduction code.

A compilation of all ten images of Saturn was attempted in order to increase our signal to noise ratio. However, due to the fact that the tracking systems of our telescopes is fickle at best, Saturn drifted on our CCD. Consequently, when a final compilation image was created, the image was dominated by noise and no distinguishable object could be seen. Therefore, each science image was reduced and stored separately.  

Lab Report - Epsilon Lyrae 1

Baker Observatory Lab Report

AST 311: Observational Techniques

Using the 8-inch telescopes at Baker Observatory

Missouri State University

Joshua W. Kern

Abstract

On September 30, 2015, we used an 8-inch telescope at Baker Observatory to take images of Saturn using the ST-I monochromatic CCD cameras without filters. Darks, flats, and sky darks were taken to reduce the science images in the future. The reduction of our science images should be completed by December.

Experiment

The purpose of this experiment is to become familiar with operating the 8-inch telescopes at Baker Observatory and obtaining reduced observational data of astronomical objects. After mounting the telescopes and plugging in the power supply, we checked the alignment of the finder scope with the telescope and focused the telescope. We then aligned the telescope to the sky using the two star align mode with Polaris and Altair. Using the preset objects programmed into the computer, we moved the telescope to view Saturn. We assured that Saturn was in the center of the field of view of the telescope and replaced the viewing lens with the ST-I monochromatic CCD camera. We took multiple test exposures to assure the telescope was in focus and then began taking science images. After obtaining 10 science images, we took 10 darks, 10 flats, and 10 sky images to use in the data reduction.

Results & Discussion

The science images we obtained of Saturn were taken with an integration time of 0.05 seconds and no filters were used. Maximum values for the counts in each image of Saturn are ~15,000 corresponding to the object itself. Two (possibly three) of Saturn's moons are visible in the images as well. An example of one science image is given below. 

 

However, this image is composed from the raw science data without corrections from the darks, flats, and sky images. Further data reduction will be needed in order to obtain data that is satisfactory for analyzing.

The 10 sky images were obtained for the purpose of correcting the noise caused by the light pollution from the sky. These were taken with the same integration time (0.05 seconds) and focus of the telescope as when the science images were being taken accepting that the images were taken of a nearly empty field of view near Saturn and not of Saturn itself. The other darks and flats were taken to correct for the instrumental errors in our data. They were taken with an integration time of 2 seconds due which was chosen due to the ability of the camera, while taking flats, to obtain reasonably high counts (>10,000). An example is given below of a dark, sky dark, and flat image respectively. 

 

This data should be sufficient to correct for instrumental and observational errors in our science images.