Wednesday, July 29, 2009
Pensacola, Florida Supervised Classification Map
This map was classified and recoded with ERDAS 9.3 using the supervised method. At least 3 training sites were picked for each land use category. Overall, it appears to be more accurate and shows the mix of residential and trees in a close up view. There are still some problems do to the pixel colors of urban and offshore currents being the same thus all being classified as urban. Also, as in the unsupervised map parts of the airport runway are either classified as urban because of the concrete or agriculture because of the color match with the farming areas. Maybe some post-processing editing would be necessary to correct these issues. There may be other tools in ERDAS that could fix the problems too or maybe more training areas would have corrected some of the issues.
Thursday, July 23, 2009
Pensacola Recoded Classification Map
The above satellite image was classified and then recoded using ERDAS 9.3. The ISODATA algorithm was used with 15 classes to be identified. From this the recode process narrowed the number of classes to 8 by combining similar classes.
Although the image roughly portrays the land characteristics on the real ground there are issues. Agricultural areas had roughly the same color as the airport runways so both were identified in the same class. Also, trees look the same whether in an urban or forest area and tree canopies tend to overwhelm parts of the urban area especially in a geographic area with prolific plant growth such as Pensacola, Florida.
These are typical problems in an unsupervised classification. One potential fix would have been to use more classes and a better color scheme to identify the different land use areas. This is primarily true for the vegetation aqnd urban areas that were difficult to differentiate from one another.
Tuesday, July 21, 2009
Image Rectification
Image rectification in GIS is the process of taking raw raster aerial photographs and transforming them into a chosen geographic coordinate system. This process is useful for cleaning up the raw raster photos and removing the distortion created during image acquisition. This distortion is caused by lens distortion, earth curvature, terrain and sensor orientation among others. A successfully rectified image will have more map like characteristics like a flat paper map. Also, if several images of the same area are rectified to the same coordinate system then "time-series" studies can be performed. If a region of photos is rectified then a mosaic can be created orientating all the images so as to create a larger area.
Drawbacks of image rectification would include; time to preprocess the map and then resample, lack of identifiable ground control points(GCP), error when choosing GCP's, image distortion and poor quality of an image that makes it difficult to identify GCP's.
Drawbacks of image rectification would include; time to preprocess the map and then resample, lack of identifiable ground control points(GCP), error when choosing GCP's, image distortion and poor quality of an image that makes it difficult to identify GCP's.
Thursday, July 9, 2009
Thermal Image
The roads, sidewalks and patios are lighter in color because they are warmer in comparison to the vegetation and other above ground features. This is because concrete on the ground holds heat more efficiently then things above ground or vegetation and this becomes apparent in thermal images taken just before dawn.
Vegetation is generally darker at night because evapotranspiration allows these items to cool at night. The variety of shades is probably because different vegetation has varying moisture content which impacts heat retention.
Storage sheds and autos are dark meaning they are cooler than the surrounding ground. Items above ground with no heat source cool quickly at night and before dawn they are the coldest they will get. Some autos have a white spot in the front indicating the hot engine of autos moved just before the image was taken.
The bright spots on the roofs of the houses are from exhaust of the heating systems. The image was taken in winter at dawn which is the coldest time of the year and the day. In summer it may have been possible to see white spots near the houses on the ground do to the AC units as they expel warm air too.
Monday, July 6, 2009
Question7 Panchromatic and Multispectral Images
Question 7
The SPOT multi-spectral false color image at the top of the page shows vegetation in red. Vegetation jumps out on the image when compared to the panchromatic image below in gray scale.
The SPOT panchromatic image appears clearer and more in focus and the white areas (sand, roads and buildings) really are contrasted compared with the darker vegetation and water areas. Also, the below ocean surfaces show greater detail.
I think the panchromatic image may be more suitable when measuring beach erosion after a storm or longshore flow given the way the sand jumps out on the image. Also, man made structures such as roads appear clearer in the panchromatic image. The multi-spectral color image appears well suited for vegetation studies both on land and possibly at the ocean surface (algae blooms, plankton etc...).
Digital Orthophoto Quadrangles
Digital Orthophoto Quadrangles (DOQ) have many positives
but there are a few negatives in using them. They include:
a small area of coverage and a straight down perspective which shows no relief or the ability to perceive depth.
DOQ's generally are used as base maps and need other layers to get much usable data.
Also, DOQ's are current only when made and my guess is they would be outdated quickly unless a new aerial photo is scanned and processed.
but there are a few negatives in using them. They include:
a small area of coverage and a straight down perspective which shows no relief or the ability to perceive depth.
DOQ's generally are used as base maps and need other layers to get much usable data.
Also, DOQ's are current only when made and my guess is they would be outdated quickly unless a new aerial photo is scanned and processed.
Subscribe to:
Posts (Atom)