Skip to Local Navigation
Skip to Content
California State University, Long Beach
CSULB Geospatial Research and Mapping (GRAM) Field Program
Print this pageAdd this page to your favoritesSelect a font sizeSelect a small fontSelect a medium fontSelect a large font
 

Daily Blog for Audrey


Day 26

Posted on June 28, 2013 by Audrey

Today I added citations to my PowerPoint. Then I asked Dr. Wechsler for improvements on my presentation. Then I continued to practice my presentation. Three people presented before me, Peter, Howard, and Cole. While they were talking I continued to practice the one slide that I kept spending three to four minutes on so that I can reduce the time to one minute. When I actually presented, I forgot about any of the recommendations that Dr. Wechlser told me and I forgot to say a lot of things and ended up talking for 6-7 minutes instead of taking 11 minutes like when I was practicing. I was glad to finally be done with the presentations, but it was kind of intimidating to talk in front of so many people.

 

Afterwards, we all went to the Haleiwa Joe’s restaurant and I had a Luau Lumpia (?). The food was really good. I cannot believe that we are going back home and I will never be coming back to the barn or the Ka’a’awa Valley. I am happy that I will be able to go back to my home where there are no barn animals making noises and most importantly, not a lot of mosquitoes. However, I feel that I will miss being in the barn with everyone.

Day 25

Posted on June 27, 2013 by Audrey

Today I worked on improving my classification of the uluhe ferns. I was able to increase the accuracy of the classification by creating one ruleset and classifying the result as extra1. Then I extracted a part of extra1 class that fulfilled another ruleset and classified that part as extra2. I kept repeating this process until I got a selection that only contains ferns then I reclassified the selection as uluhe ferns. I did the same for trees, and grass.

I also started my PowerPoint slides and organizing my data for the data dump. I am worried whether I will have enough time to work on the paper tomorrow, but I am planning on working on the paper and the metadata tomorrow. For today, I am going to continue editing my PowerPoint slides and practice presenting.

 

Day 24

Posted on June 26, 2013 by Audrey

Today I talked with Briton and I will be trying to use a slightly different way of classifying the ferns to make the classification more accurate. Up until now, I had been setting an upper and lower limit for a single variable to classify an object as fern. But now, I am working on finding two variables that can be bounded to extract the ferns more accurately. So far, I haven’t had much luck but hopefully I will be able to finish this tonight.

 

I have started working on my PowerPoint presentation as well. Hopefully, I will be able to finish the classification tomorrow morning and start spending some time on my presentation and the written portion of my project.

Day 23

Posted on June 25, 2013 by Audrey

Today I worked on extracting the trees from the ferns. I finally decided to write down all of the mean values for each object that I want to classify as ferns. I found that all of the numbers are fairly far apart from each other and even if they are close, the range is still wide enough to include lots of unwanted objects in the classification. I was surprised to see that the grass was extracted pretty easily but the trees are the most difficult to extract because the range of values covered by trees is similar to the range covered by the ferns.

 

I am going to continue trying to extract the trees but I will also start preparing for my presentation and work on the poster tomorrow. I hope that I can come up with a better solution tomorrow.

 

Day 22

Posted on June 24, 2013 by Audrey

Today I stayed in the barn the whole day and worked on classifying the ferns using eCognition. I chose to work with objects segmented at a scale of 50, compactness of 0.9, and shape of 0.1. I think that this makes the objects small enough to make each object contain only grass or only trees rather than a mix of both. I started out using Hue to select almost everything in the valley and classified them as ferns. Then I reclassified parts of it as grass. I have classified shadows as shadows as well. I am now trying to extract the trees from the ferns. However, the values are so similar that it is difficult to do.

 

The points I have collected were not very useful because most of them lie in the middle of trees without any fern or grass-like patches nearby. It seems that the places where ferns grew have increased since 2011, which is when the WorldView 2 imagery was taken.

 

Gordon, Greg, and I were responsible for cooking today. I ended up just making the salad and cutting up the bread, but the food was great. When I was washing the dishes, I realized that I had small cuts all over my hand from yesterday’s little vacation. I probably got them from accidentally hitting the corals as I attempted to swim past them.

Day 21

Posted on June 23, 2013 by Audrey

Today I woke up at 4:58am and started to pack up so that I could leave by 5:15am. I ended up getting ready to leave until 5:30am though. I went with Paul, Michelle, Gordon, Cole, Thomas, and James. This was so that we could go to the Hanauma Bay before the place gets packed with people. I saw a lot of fish with very vibrant colors. I also tried to take some pictures of the fish using Michelle’s camera, but it was difficult because the fish were moving pretty fast.

 

Afterwards, we went to eat lunch at Teddy’s Bigger Burgers and I ate a 5oz Hawaiian Style burger. The burger was huge and expensive but it was really good. The meat seemed like the kind of meat that was grilled at the barn rather than the kind you get at a McDonald’s.

 

After enjoying a very fulfilling meal, we went to Pearl Harbor. I took some pictures and went into the gift shop to buy some gifts but I was not able to get gifts for everyone. Today was a great break from working on eCognition at the barn.

Day 20

Posted on June 22, 2013 by Audrey

I was working on trying to improve my ferns classification and ended up losing the file that contained the classification. I am probably just going to continue searching for a way to improve the ferns classification rather than going back and trying to figure out what I did before. I have tried too many variations of the same classification that I don’t really want to go through that process all over again.

 

From now on, I will definitely keep track of which classification methods were successful and how I did them. I am currently trying to read more articles to see if there are different ways of classifying vegetation other than just comparing mean values of the bands and slope.  I am hoping to find some solutions that I am capable of doing.

Day 19

Posted on June 21, 2013 by Audrey

Today I worked on trying to load the slope data into eCognition and finally managed to bring it in with Briton’s help. After that I finally started to do some classifications using the slope.

 

I noticed that the objects classified as ferns were at the back of the valley and there were more on the left side of the valley or the side that does not have Atlantis, which was something I noticed when I was in Ka’a’awa Valley two days ago. So I knew that in general, using the slope worked pretty well, except for the fact that each object was so big that it included trees, ferns, and grass. Also, the program classified a road as ferns as well which was weird. So then, I tried to decrease the size of each object when the World View 2 imagery was segmented by changing the scale parameter and the compactness value. By changing the compactness value to 0.9 instead of the default 0.5 and having the scale parameter 100, each object seemed to be more accurate. I found that except for 2 objects everything was pretty much at the locations where I logged the GPS points or at locations I expected to see ferns because of what I remember about the unreachable ferns when I was at the valley a couple of days ago.

 

I will probably work on decreasing the scale parameter by 10s and see what fits better. I will also try to use the actual values of the pixel in WV2 imagery to eliminate the misclassified objects but I am still not sure about how to grab the couple of ferns that have not been classified as ferns at any point during my classification process. However, after managing to get the slope raster into eCognition, I am finally starting to enjoy using eCognition. Its ability to classify raster imagery is amazing.

 

I don’t think I need to go out to the field ever again since the number of fern patches in the valley is very limited and I have no reason to visit the site again since I remember what they looked like from close up as well as how they looked from the road on the opposite side of the valley.

Day 18

Posted on June 20, 2013 by Audrey

Today I went to the Ka’a’awa Valley but once I arrived at the base station, I tried using the spectrometer and I realized this might not work out so well. I would have to hold the Yuma in one hand and the spectrometer on the other and take the spectral reading while standing on a steep slope and on top of the ferns. Later, I talked with Dr. Lee who said that I probably would not have to measure the slope since I could just get the approximate slope using the DEM. Since I already collected all the GPS points for the ferns that are reachable, I came back to the barn.

 

Once I returned to the barn, I started working on adding the field notes to the GPS points. Then I used eCognition to try to do some classifications. But I had a lot of trouble getting the DEM to appear on ArcMap and eCognition. The trouble mostly arose from the fact that the DEM was in asc or adf format. I am still unsure whether I am opening the DEM correctly. After getting the DEM to appear in ArcMap, I created a slope raster which is also composed of adf files.

Day 17

Posted on June 19, 2013 by Audrey

Today I tried to use eCognition to use the World View 2 imagery and do some classifications before I went out into the field. I am still really slow at using eCognition but hopefully I will improve at using it quickly.

 

Once I arrived in the Ka’a’awa Valley, I was dropped off at Atlantis and I started collecting points for ferns and grass. I used a Garmin GPSmap 60C5x which just takes GPS points with a unique ID. I took notes on whether the points were grass or ferns. I only ended up covering the backside of the valley because that seemed to be the only places where the ferns existed that was reachable and unreachable. I noticed that the ferns seemed to prefer the side of the Valley that is opposite to Atlantis. I collected the grey, brown, and green portion of the ferns because I recognized something high up on the slope as ferns rather than grass because of the brown mixed in with green so I thought that the imagery might catch the differences as well.

 

Once I got back to the barn, I helped a little with the cooking because I wanted to see how to make the stir fried chicken and vegetables. I also read the instructions for the spectrometer so that I won’t have too much trouble using it out in the field tomorrow. I also asked about how to measure the slope, which can apparently be done using a clinometer. I am hoping to use that as well tomorrow and finish up the data collection tomorrow.

About Audrey

I'm a senior at University of North Carolina at Chapel Hill and I am majoring in Geography with a concentration in GIS.