DRAFT: This module has unpublished changes.

Week 1

The opening passage of my "Statement of Purpose" during the application process to Northeastern University. 


         Unlike the other kids in high school and college who were focused on sports and extra-curricular activities, I could always be found in a library with my head buried in a news article or National Geographic. I have always been fascinated with understanding how organized crime and terrorism affect geopolitical relations, and what can be done to counter it. Before I even knew what GIS/Remote Sensing was, I was already interested with the applications of satellite imagery with photographic analysis and how it could be used to predict and stop crime.

       During my Criminal Justice studies at the University of Scranton, I learned that in today's globalized world, many criminal organizations have adapted by becoming multifaceted, investing themselves in a broader range of illegal activities. Choosing to focus on both Central America and the Middle East, I took courses immersing myself in regional history, culture, politics, even going as far as studying both the Spanish and Arabic languages. I familiarized myself with the issues plaguing both regions and have noticed a common thread. Building off the research of Australian National University’s Ms. Julie Ayling's theory of “Criminal Organizations and Resilience”, I developed my own personal theory. Ms. Ayling proposes that in order for criminal organizations to survive pressures from local law enforcement and competing criminal enterprises, it must adapt new practices to resist and overcome (Ayling, 2009, p. 183). I go a step further and suggest that there are primarily three factors in an environment that will determine whether or not a criminal organization is capable of adapting. These three factors are as long as a group is in an environment that has access or the ability to conceal its illicit activity, find an alternative financial resource, or receives some type of foreign state sponsorship, the more likely a criminal organization can adapt, reestablish, and continue operations. Essentially, this is the natural cycle of how criminal organizations stay relevant or collapse in the undercurrent of society.


                         My mission is to break this cycle.


(Ayling, 2009, p. 183)

Ayling, J. (2009). Criminal organizations and resilience (37th ed., Vol. 4, International Journal of Law Crime and Justice). Retrieved February 19, 2017, from https://www.researchgate.net/publication/228414359_Criminal_organizations_and_resilience


I have a passion for geopolitical issues and have an extensive background in the cultural, economical, and historical factors that impact our daily lives. During my high school and college years, I struggled finding an unifying field of study that allowed me to bridge these subjects together. When I discovered the field of Remote Sensing and GIS, I rejoiced because I knew I found where I belonged. I had finally found a field that gave me the tools to make a change.            


In this week’s first discussion post, I was so confused and lost. I’m not the greatest with computers, but I enjoyed taking raw internet data and experimenting with it. Not 100% sure with how I did, but I was able to create an image of what appears to be hurricane Katerina charging into New Orleans!





DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.

Week 2


Geometric Correction


This week we focused on “Geometric Errors”, or distortions that occur during the process of obtaining information. Satellite imagery is collected in large “ground swaths” running North to South. As the image is being gathered, there are a series of issues that can skew the data. Due to the fact that the Earth is not a true flat surface nor a perfect sphere, issues such as 1 Dimensional Relief Displacement and Tangential Scale Distortion occur.


DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
Radiometric Correction
For week 3, I found myself falling falling behind the rest of the class.
I was having a hard time understanding which unique frequncies and color bands of light energy produce certain spectral images.

Fortunately, in my Fundamentals of Remote Sensing class, the I was directed to a particular website. This website, provided a wide range of pre-made images using the RGB color guns. In addition, it was set up to mimic real satellites, Landsat being one of them. I spent at least an hour experimenting when I decided to hand copy the entire list of Bands and their associated traits. I copied it on to my phone, and refer to it whenever I am using Landsat imagery (or equilivant). 


Band 1 (0.45-0.52um, blue-green):

Since this short wavelength of light penetrates better than the other bands, it is often the band of choice for aquatic ecosystems. It is used to monitor sediment in water, mapping coral reefs, and water depth. Unfortunately, this is the noisiest of the Landsat bands since short wavelength blue light is scattered more than the other bands. For this reason, it is rarely used for “pretty picture” type images.


Band 2 (0.52-0.60um, green):

This has similar qualities to band 1, but not as extreme. This band is often selected because it matches the wavelength for the green we seen when looking at vegetation.


Band 3 (0.63-0.69um, red):

Since vegetation absorbs nearly all red light (it is often referred to as the “Chlorophyll absorption” band), this band can be useful for distinguishing between vegetation, soil, and with monitoring vegetation health.


Band 4 (0.76-0.90um, near infrared):

Since water absorbs nearly all light at this wavelength, bodies of water appear very dark. This contrast with bright reflectance for soil and vegetation, making it a good band for defining the water/land interface.


Band 5 (1.55-1.75um, mid-infrared):

This band is very sensitive to moisture, and therefore is used to monitor vegetation and soil moisture. It is also good at differentiating between clouds and snow.


Band 6 (10.40-12.50um, thermal infrared):

This is a thermal band, which means it can be used to measure surface temperature. This is primarily used for geological applications, but it is sometimes used to measure plant heat stress. This is also used to differentiate clouds from bright spots, since clouds tend to be very cold. One other difference between this band and the other multispectral ETM bands, it that the resolution is half the of the other bands (60m instead of 30m).


Band 7 (2.08-2.35um, mid infrared)

This band is also used for vegetation moisture, although generally band 5 is preferred for that application, as well as for soil and geology mapping.


Landsat Band Combinations (Suggestions)

R=3 G=2 B=1 or 321

This color composition s as close to true that can be obtained with Landsat. Very useful for studying aquatic habitats. Downside is that images can be hazy.

R=4 G=3 B=2 or 432

This has similar qualities to the image with bands 3,2,1,-- however, since this includes the near infrared channel (band 4), land—water boundaries are clearer and different types of vegetation are more apparent. This was a popular band combination for Landsat MSS data since that it did not have a mid-infrared band

R=4 G=5 B=3 or 453

This is crisper than the previous two images because the two shortest wavelength bands (bands 1 and 2) are not included. Different vegetation types can be more clearly defined and the land/water interface is very clear. Variations in moisture content are evident with this set of bands. Very popular.

R=7 G=4 B=2 or 742

This has similar properties to the 453 combo, but the biggest difference is that vegetation is green.


I also wanted to memorize the Electromagnetic Spectrum, so I developed a quick story.

Blue to red are growing lengths
 NIR-- -- -- - 1.2µm
 Red-- -- -- - 0.7µm
 Green-- -- 0.6µm
 Blue-- -- -- 0.4µm
 Ultraviolet—0.3µm

Story to memorize EMS

Violet was years 3 old when she started
picking BLUE flowers at 4 in the morning.
Green with envy, 6 girls teased her. Her face
Red, she shouted 7 times at them. NIR by, a
clock struck 12.
DRAFT: This module has unpublished changes.

                                                      Week 4

             So recently, I found some interesting infrastructure in the rainforests of Central America. While I absolutely wanted to share it with my fellow students, being that I have loved ones in that region of the world, its best to keep my mouth shut and head down. Despite that fact, I want to continue my personal studies and implement whatever skills I learn in class on this subject. I just simply will NOT elaborate on what it is I’m looking at.

            After this week’s Egypt assignment, I was simply amazed by the power of temporal data. While I wanted to sit down and experiment further with the techniques demonstrated in this assignment, my work schedule made it impossible for me to access the Northeastern Remote Server, so I turned to alternative means of temporal satellite data. Three in particular were Google Maps, Planet.com, and ChangeMatters.


While I'm not going to show every single temporal image, I was able to look at the area over time. What I have gather is this-


1) ChangeMatters has demonstrated that this region first appeared between 2005 and 2010.

2)Planet.com has demonstrated that earlier this year, it looked ike the region had been cut down again.

3)Google maps is showing that the area current seems to be abadoned.


At this point, I'm not sure if this is anything, but regardless, this has been an interesting process. Just before throwing in the towel, I used my theory of process of elmination, or the classes when spoke about in class, and I found the last image. I'm not sure of what that is, but that looks like an strip, is in the middle of nowhere, and has no clear and obvious roads leading to it.


DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.
User-uploaded Content
DRAFT: This module has unpublished changes.

Week 5


Unsupervised Classification


Carrying over from the work of the previous week, I decided to apply the spectral clusterring software to a recent image of Peten, Guatemala. 

I know that going forward, a combination of both pattern recognition and spatial recognition will important indentifying potential sites.

I have no images for the week's findings, due to software issues, but I know what the complications are.


In order to locate hidden infrastructure that services small aircraft, I think that using a pratice/training image of a regular airport in important in helping the computer locate target sites. The issue that the software had was that this region of Central America has numerous farms that are irregularlly shaped. This of course becomes noise as the computer tries to sort out the data into different classifications. 


To safely (hahaha I'm sure they cannot call OSHA for guidance) land an aircraft, certain dimensions and elevations are neccessary. If I can program that into the software, then perhaps I'll have more precise results. 





DRAFT: This module has unpublished changes.



I’m beginning to concede to the truth that I’m a very stubborn person who refuses to think in terms of spectral applications, and only sees the world through a spatial mindset.

So, the fact that I had to teach the ENVI supervised classification filter to target which groupings of pixels, I was forced somewhat to think in terms of spectral resolutions.    

Thank goodness for “Established Classification Schemes”!

In all serious, this was an informative week that brought to my attention some new possibilities. While I might be able to locate areas that fit certain spatial/logical filters, it would be wise to use spectral images techniques to help bolster or back up my suspicions.

Whether it’s the presence of drug crops or an absence of jungle vegetation, there’s bound to be a pattern that I can isolate that will help narrow down my target area. Using spectral analysis to help refine and refocus my search area is crucial.

While ENVI may not be may favorite GIS/RMS platform, I am beginning to see the powerful potential it has.

Unfortunately, this week I have decided to stop using rainforest imagery because I keep downloading the wrong information. While it was more interesting to apply my own personal area of interest, I’m actual having more success with using the provided material.

DRAFT: This module has unpublished changes.
DRAFT: This module has unpublished changes.



I will try to provide a recap of this week’s material.

Principal Component Analysis (PCA), is a type of image transformation that reduces redundancy within a dataset. By compressing redundant information, variations within the dataset are highlighted.

IF you have 7 thematic bands, you will to generate 7 PCAs, and pick the ones with the most variation.

In terms of visualizing a PCA, one can use them to detect changes in an image.

I have provided a link to a video that best helps to describe this process.




DRAFT: This module has unpublished changes.

Week 9 Data Integration


This week, I got my first chance to test out actual DEMs. Being that I have a variety of different interests, I have been experimenting with the software on ArcGIS Scene.

My most ambitious project so far has been to recreate the region of Guatemala my family comes from. I the following image gallery, I provide imagery of the various stages have gone through.

In the spirt of data integration, my DEM is mixture of elevation data, shapefiles, natural color satellite imagery, and some artistic vertical exaggeration.

DRAFT: This module has unpublished changes.
DRAFT: This module has unpublished changes.

Week 10: Mosaics

This week has been very interesting. As I have continued on with my DEM experiments, I have found that continuously having to juggle various datasets is very frustrating. The idea of establishing a “synoptic” or expanded map or DEM would be very helpful.

Unfortunately, it is not working as it is in the class assignment.

I’ve been having issues get to transfer data from the Remote server ENVI 32 bit, and because my personal ENVI is 64 bits, it will not transfer data over to my ArcGIS programs. So, as a result I have been trying to use the mosaic features on the ArcGIS Scene and Map, but apparently, they cannot read the data.

In a perfect world, I would like to take the DEM data and combine it with adjacent data to create one large synoptic view (perhaps there is not enough overlap to establish a successful mosaic) from which I can intersect and clip out my target areas.  

Once I have established my target DEM region (having now reduced the size of spatial data), I would want to start experimenting with various Spectral filters that I can drape over the DEM.

 The last few images are of a successful draping of a satellite image on top of Hawaii.

DRAFT: This module has unpublished changes.
DRAFT: This module has unpublished changes.



In my Fundamental of Remote Sensing Class, I decided to work on a DEM of the surface of Mars, and while I did my very best to try and preserve the spatial and spectral resolution, I found myself in a bind.

So, I began to experiment will some of the contrast stretches and spatial features to see if I could increase some of the spatial resolution.

In the last picture, I applied a stretched contrast, applied the hill shade function, and used a Minimum- Maximum spatial filter.  

There is a stark difference between the original image and the final version. I believe that I was very successful in altering the DEM to vastly improve the spatial resolution. 

DRAFT: This module has unpublished changes.
DRAFT: This module has unpublished changes.