Engineering Commons^™

Investigation Of Green Strawberry Detection Using R-Cnn With Various Architectures, Daniel W. Rivers

Master's Theses

Traditional image processing solutions have been applied in the past to detect and count strawberries. These methods typically involve feature extraction followed by object detection using one or more features. Some object detection problems can be ambiguous as to what features are relevant and the solutions to many problems are only fully realized when the modern approach has been applied and tested, such as deep learning.

In this work, we investigate the use of R-CNN for green strawberry detection. The object detection involves finding regions of interest (ROIs) in field images using the selective segmentation algorithm and inputting these regions …

Visual Speech Recognition Using A 3d Convolutional Neural Network, Matthew Rochford

Master's Theses

Main stream automatic speech recognition (ASR) makes use of audio data to identify spoken words, however visual speech recognition (VSR) has recently been of increased interest to researchers. VSR is used when audio data is corrupted or missing entirely and also to further enhance the accuracy of audio-based ASR systems. In this research, we present both a framework for building 3D feature cubes of lip data from videos and a 3D convolutional neural network (CNN) architecture for performing classification on a dataset of 100 spoken words, recorded in an uncontrolled envi- ronment. Our 3D-CNN architecture achieves a testing accuracy of …

Atrengine: An Orientation-Based Algorithm For Automatic Target Recognition, Justin Ting-Jeuan Kuo

Master's Theses

Automatic Target Recognition (ATR) is a subject involving the use of sensor data to develop an algorithm for identifying targets of significance. It is of particular interest in military applications such as unmanned aerial vehicles and missile tracking systems. This thesis develops an orientation-based classification approach from previous ATR algorithms for 2-D Synthetic Aperture Radar (SAR) images. Prior work in ATR includes Chessa Guilas’ Hausdorff Probabilistic Feature Analysis Approach in 2005 and Daniel Cary’s Optimal Rectangular Fit in 2007.

A system incorporating multiple modules performing different tasks is developed to streamline the data processing of previous algorithms. Using images from …

Synthetic Aperture Radar Imaging Simulated In Matlab, Matthew Schlutz

Master's Theses

This thesis further develops a method from ongoing thesis projects with the goal of generating images using synthetic aperture radar (SAR) simulations coded in MATLAB. The project is supervised by Dr. John Saghri and sponsored by Raytheon Space and Airborne Systems. SAR is a type of imaging radar in which the relative movement of the antenna with respect to the target is utilized. Through the simultaneous processing of the radar reflections over the movement of the antenna via the range Doppler algorithm (RDA), the superior resolution of a theoretical wider antenna, termed synthetic aperture, is obtained. The long term goal …

Target Tracking Using Various Filters In Synthetic Aperture Radar Data And Imagery, Jessica L. Kiefer

Master's Theses

This thesis explores the use and accuracy of several discrete-time image filters for the purpose of target tracking in Synthetic Aperture Radar imagery. Both extended targets and point targets are used for tracking, showing the need for different types of filters for each target type.

Monte Carlo analysis is performed on the results of the extended target filter results to determine the absolute mean-squared error between the filter prediction of the target centroid and the actual location of the target centroid. Two different filters were chosen for the extended target: Kalman and H Infinity.

Both the Kalman and H Infinity …