Master theses projects

1. Algorithms for registration of gel images produced in 2D electrophoresis experiments
   Student: Anna Persson
   Supervisor: Lennart Björkesten, Amersham Biosciences, Uppsala
   Examiner: Ingela Nyström
   Publisher: CBA Master Thesis Report No. 53, 36p., 2002,
   Uppsala University School of Engineering, UPTEC F 02 011
   Abstract: The standard method for separating proteins is two-dimensional gel electrophoresis (2DGE). The subsequent data analysis of the obtained digital gel images, is a bottleneck in the 2DGE procedure. One step in this analysis involves matching of protein patterns between gels, where proteins are visualised as spots. In existing matching methods, the images are normally matched after protein spots in the different gel images have been detected and assigned positional coordinates and identifying numbers. By reducing the gel images into a list of spot locations, a considerable amount of visual information, such as shape and intensity spread, is lost. Hence, matching algorithms that incorporate all available image information throughout the process are preferable. Such algorithms are commonly known as warping algorithms. Existing available warping algorithms have been investigated, and a suitable approach has been selected for a prototype implementation.
   The implemented algorithm uses a non-parametric transformation called piecewise bilinear mapping (PBM) together with a multi-resolution approach, which speeds up the optimisation process and removes misleading local optima close to the global optimum. A correlation function is used as a similarity measure and optimised using the BFGS algorithm. To evaluate the algorithm, images taken using the difference gel electrophoresis (DIGE) technique were used. This technique is based on fluorescence dyes. The testing of the warping algorithm on DIGE image pairs has produced successful results and the method has a potential to become a helping tool for existing matching methods. Some aspects of possible future improvements have also been identified.

2. Implementation and evaluation of image analysis based seed classification and sorting system
   Student: Torgil Svensson
   Supervisor: Jaan Luup, Nekros AB, Uppsala
   Examiner: Ewert Bengtsson
   Publisher: CBA Master Thesis Report No. 54, 43p., 2002,
   Uppsala University School of Engineering, UPTEC IT 02 021
   Abstract: An automatic seed sorting and classification system was implemented and evaluated. Image processing hardware was assembled and software was developed. Sorting divided 850 g seed samples into two fractions containing typical and non-typical seeds. Misclassification, when analysing a seed sample of a certain species, was defined as sorting seed kernels from other species into the fraction of typical seeds. Classification was based on statistical analysis of 22 morphological, colour and texture features extracted from seed kernel images. A decision algorithm, based on linear discriminant analysis and Mahalanobis distances, was used. Seed samples from eleven varieties of rye, barley, triticale, wheat and oats were classified into five classes. Classification into six and eleven classes was evaluated and a lower accuracy than with five classes was found. Canonical discriminant analysis was performed. Extracted features were arranged in order of significance by stepwise discriminant analysis. Maximum misclassification of 2% was found for classification from and into rye, barley, wheat and oats. Misclassification of barley and oats into triticale was 2.5% and 4.1%, respectively. Results were achieved by sorting less than 10% of the sample as non-typical seeds. Requirements set for the project were met and the system will be developed into a commercial product.

3. Image analysis of cast-iron
   Student: Roya Khomand
   Supervisor: Jan Linder, Scania, Södertälje
   Examiner: Gunilla Borgefors
   Publisher: CBA Master Thesis Report No. 55, 62p., 2002,
   Scania, Reg. No. 02-580
   Abstract: This master thesis discusses the possibility of distinguishing salient features of cast-iron using a commertial image analysis system.
   There are three different types of cast-iron, ``Ductile iron'', ``Grey iron'', and ``Compacted graphite iron''. It is the form of microstructure of graphite that determines the different types of cast-iron.
   Distinguishing cast-iron is presently done in accordance with the SS-EN ISO 945 standard that is visual comparison between reference photos and the test samples. Since the standard consists of visual manual comparison, there is the risk that the ascertainment, to a certain extent may be dependent on the operator. In order to distinguish the composition of cast-iron more objectively, as well as to allow more room for a number of measurable parameters for cast-iron, image analyse could be a powerful tool. In this study, it was possible to use image analyse as a tool to distinguish the different types and size of the graphite particles, but not their distribution. (In Swedish)

4. Road sign recognition from a moving vehicle
   Student: Björn Johansson
   Supervisor and Examiner: Ewert Bengtsson
   Publisher: CBA Master Thesis Report No. 56, 71p., 2002
   Abstract: This project aims to research the current technology for recognising road signs in real-time from a moving vehicle. The most promising technology for intelligent vehicle systems is vision sensors and image processing, so this is examined the most thoroughly. Different processing algorithms and research around the world concerned with sign recognition are investigated. A functioning system has also been implemented using standard web-camera mounted in a testing vehicle. The system is restricted to speed signs and achieves good performance thanks to fast but still robust algorithms. Colour information is used for the segmentation and a model matching algorithm is responsible for the recognition. The human-computer interface is a voice saying what sign has been found.