A study of the soil characteristics, weather conditions, and effect of management skills on the yield of the agricultural crop requires site-specific details, which involves large amount of labor and resources, compared to the traditional whole field based analysis. This thesis discusses the design and implemention of yield monitor for sweetpotatoes grown in heavy clay soil. A data acquisition system is built and image segmentation algorithms are implemented. The system performed with an R-Square value of 0.80 in estimating the yield. The other main contribution of this thesis is to investigate the effectiveness of statistical methods and neural networks to correlate image-based size and shape to the grade and weight of the sweetpotatoes. An R-Square value of 0.88 and 0.63 are obtained for weight and grade estimations respectively using neural networks. This performance is better compared to statistical methods with an R-Square value of 0.84 weight analysis and 0.61 in grade estimation.

No description available.

No description available.

No description available.

No description available.

No description available.

This work implements and evaluates different methods of image analysis and manipulation for the purposesof object recognition. It lays the groundwork for possible future projects that could use machine learning onthe output for the purposes of analyzing the behaviour of lab mice. Three different methods are presented,implemented on a selection of examples and evaluated.

This master thesis investigates different approaches for integrating prior information into a neural network for segmentation of medical images. In the study, liver and liver tumor segmentation is performed in a cascading fashion. Context channels in the form of previous segmentations are integrated into a segmentation network at multiple positions and network depths using different integration strategies. Comparisons are made with the traditional integration approach where an input image is concatenated with context channels at a network’s input layer. The aim is to analyze if context information is lost in the upper network layers when the traditional approach is used, and if better results can be achieved if prior information is propagated to deeper layers. The intention is to support further improvements in interactive image segmentation where extra input channels are common. The results that are achieved are, however, inconclusive. It is not possible to differentiate the methods from each other based on the quantitative results, and all the methods show the ability to generalize to an unseen object class after training. Compared to the other evaluated methods there are no indications that the traditional concatenation approach is underachieving, and it cannot be declared that meaningful context information is lost in the deeper network layers.

Image layers are two-dimensional planes, each comprised of objects extracted from a two-dimensional (2D) image of a scene. Multiple image layers together make up a given 2D image, similar to the way a stack of transparent sheets with drawings together make up a scene in an animation. Extracting layers from 2D images continues to be a difficult task. Image compositing is the process of superimposing two or more image layers to create a new image which often appears real, although it was made from one or more images. This technique is commonly used to create special visual effects in movies, videos and television broadcast. In the widely used "blue screen" method of compositing, a video of a person in front of a blue screen is first taken. Then the image of the person is extracted from the video by subtracting the blue portion in the video, and this image is then superimposed on to another image of a different scene, like a weather map. In the resulting image, the person appears to be in front of a weather map, although the image was digitally created. This technique, although popular, imposes constraints on the object color and reflectance properties and severely restricts the scene setup. Therefore layer extraction and image compositing remains a challenge in the field of computer vision and graphics. In this research, a novel method of layer extraction and image compositing is conceived using a moving-aperture lens, and a prototype of the system is developed. In an image sequence captured with this lens attached to a standard camera, stationary objects in a scene appear to move. The apparent motion in images is created due to planar parallax between objects in a scene. The parallax information is exploited in this research to extract objects from an image of a scene, as layers, to perform image compositing. The developed technique relaxes constraints on object color, properties and requires no special components in a scene to perform compositing. Results from various indoor and outdoor stationary scenes, convincingly demonstrate the efficacy of the developed technique. The knowledge of some basic information about the camera parameters also enables passive range estimation. Other potential uses of this method include surveillance, autonomous vehicle navigation, video content manipulation and video compression. / Ph. D.

This dissertation concerns the development of an interactive machine learning method for refinement and analysis of segmented computed tomography (CT) images. This method uses higher-level domain-dependent knowledge to improve initial image segmentation results. A knowledge-based refinement and analysis system requires the formulation of domain knowledge. A serious problem faced by knowledge-based system designers is the knowledge acquisition bottleneck. Knowledge acquisition is very challenging and an active research topic in the field of machine learning and artificial intelligence. Commonly, a knowledge engineer needs to have a domain expert to formulate acquired knowledge for use in an expert system. That process is rather tedious and error-prone. The domain expert's verbal description can be inaccurate or incomplete, and the knowledge engineer may not correctly interpret the expert's intent. In many cases, the domain experts prefer to do actions instead of explaining their expertise. These problems motivate us to find another solution to make the knowledge acquisition process less challenging. Instead of trying to acquire expertise from a domain expert verbally, we can ask him/her to show expertise through actions that can be observed by the system. If the system can learn from those actions, this approach is called learning by demonstration. We have developed a system that can learn region refinement rules automatically. The system observes the steps taken as a human user interactively edits a processed image, and then infers rules from those actions. During the system's learn mode, the user views labeled images and makes refinements through the use of a keyboard and mouse. As the user manipulates the images, the system stores information related to those manual operations, and develops internal rules that can be used later for automatic postprocessing of other images. After one or more training sessions, the user places the system into its run mode. The system then accepts new images, and uses its rule set to apply postprocessing operations automatically in a manner that is modeled after those learned from the human user. At any time, the user can return to learn mode to introduce new training information, and this will be used by the system to updates its internal rule set. The system does not simply memorize a particular sequence of postprocessing steps during a training session, but instead generalizes from the image data and from the actions of the human user so that new CT images can be refined appropriately. Experimental results have shown that IntelliPost improves the segmentation accuracy of the overall system by applying postprocessing rules. In tests two different CT datasets of hardwood logs, the use of IntelliPost resulted in improvements of 1.92% and 9.45%, respectively. For two different medical datasets, the use of IntelliPost resulted in improvements of 4.22% and 0.33%, respectively. / Ph. D.