Matching Items (2)
Filtering by

Clear all filters

154717-Thumbnail Image.png
Description
Large datasets of sub-meter aerial imagery represented as orthophoto mosaics are widely available today, and these data sets may hold a great deal of untapped information. This imagery has a potential to locate several types of features; for example, forests, parking lots, airports, residential areas, or freeways in the imagery.

Large datasets of sub-meter aerial imagery represented as orthophoto mosaics are widely available today, and these data sets may hold a great deal of untapped information. This imagery has a potential to locate several types of features; for example, forests, parking lots, airports, residential areas, or freeways in the imagery. However, the appearances of these things vary based on many things including the time that the image is captured, the sensor settings, processing done to rectify the image, and the geographical and cultural context of the region captured by the image. This thesis explores the use of deep convolutional neural networks to classify land use from very high spatial resolution (VHR), orthorectified, visible band multispectral imagery. Recent technological and commercial applications have driven the collection a massive amount of VHR images in the visible red, green, blue (RGB) spectral bands, this work explores the potential for deep learning algorithms to exploit this imagery for automatic land use/ land cover (LULC) classification. The benefits of automatic visible band VHR LULC classifications may include applications such as automatic change detection or mapping. Recent work has shown the potential of Deep Learning approaches for land use classification; however, this thesis improves on the state-of-the-art by applying additional dataset augmenting approaches that are well suited for geospatial data. Furthermore, the generalizability of the classifiers is tested by extensively evaluating the classifiers on unseen datasets and we present the accuracy levels of the classifier in order to show that the results actually generalize beyond the small benchmarks used in training. Deep networks have many parameters, and therefore they are often built with very large sets of labeled data. Suitably large datasets for LULC are not easy to come by, but techniques such as refinement learning allow networks trained for one task to be retrained to perform another recognition task. Contributions of this thesis include demonstrating that deep networks trained for image recognition in one task (ImageNet) can be efficiently transferred to remote sensing applications and perform as well or better than manually crafted classifiers without requiring massive training data sets. This is demonstrated on the UC Merced dataset, where 96% mean accuracy is achieved using a CNN (Convolutional Neural Network) and 5-fold cross validation. These results are further tested on unrelated VHR images at the same resolution as the training set.
ContributorsUba, Nagesh Kumar (Author) / Femiani, John (Thesis advisor) / Razdan, Anshuman (Committee member) / Amresh, Ashish (Committee member) / Arizona State University (Publisher)
Created2016
Description

In the spring of 2016, the City of Apache Junction partnered with the School of Geographical Sciences and Urban Planning at Arizona State University on three forward-thinking plans for development in Apache Junction. Graduate students in the Urban and Environmental Planning program worked alongside City staff, elected officials and the

In the spring of 2016, the City of Apache Junction partnered with the School of Geographical Sciences and Urban Planning at Arizona State University on three forward-thinking plans for development in Apache Junction. Graduate students in the Urban and Environmental Planning program worked alongside City staff, elected officials and the public to identify opportunities and visions for:
       1. Multi-modal access and connectivity improvements for City streets and open space.
       2. Downtown development.
       3. A master-planned community on state land south of the U.S. 60.

The following sections of the report present Apache Junction’s unique characteristics, current resident demographics, development needs and implementation strategies for each project:
       1. Community Profile
       2. Trail Connectivity Master Plan
       3. Downtown Visioning
       4. State Land Visioning

The Trail Connectivity Master Plan optimizes existing trails and wide road shoulders to improve multi-modal connections across the city. The proposed connections emphasize access to important recreation, education and other community facilities for pedestrians, equestrians and bicycles. Trail and lane designs recommend vegetated buffers, wherever possible, to improve traveler safety and comfort. The proposals also increase residents’ interaction with open space along urban-rural trails and park linkages to preserve opportunities to engage with nature. The objectives of the report are accomplished through three goals: connectivity, safety improvements and open space preservation.

Downtown Visioning builds on a large body of conceptual design work for Apache Junction’s downtown area along Idaho Road and Apache Trail. This report identifies three goals: to establish a town center, to reestablish the grid systems while maintaining a view of the Superstition Mountains, and to create an identity and sense of place for the downtown.

State Land Visioning addresses a tract of land, approximately 25 square miles in area, south of the U.S. 60. The main objective is to facilitate growth and proper development in accordance with existing goals in Apache Junction’s General Plan. This is accomplished through three goals:
       1. Develop a foundation for the creation of an economic corridor along US-60 through
           preliminary market research and land use planning.
       2. Create multi-modal connections between existing development north of US-60 and
           future recreational space northeast of US-60.
       3. Maintain a large ratio of open space to developed area that encompasses existing
           washes and floodplains using a master planned community framework to provide an
           example for future land use planning.