P Slosarz

Neural image analysis for estimating aerobic and anaerobic decomposition of organic matter based on the example of straw decomposition

The purpose of the project was to identify the degree of organic matter decomposition by means of a neural model based on graphical information derived from image analysis. Empirical data (photographs of compost content at various stages of maturation) were used to generate an optimal neural classifier (Boniecki et al. 2009, Nowakowski et al. 2009). The best classification properties were found in an RBF (Radial Basis Function) artificial neural network, which demonstrates that the process is non-linear.

Classification of chosen orchard pests using the self- organizing feature maps neural network

The Kohonen neural networks are modelled on the topological properties of the human brain. These networks are also known as self-organizing feature maps (SOFM). One advantage of suggesting a procedure is the ability of the SOFM neural network to determine the degree of similarity occurring between classes. The SOFM network can also be used to detect regularities occurring in the obtained empirical data. If at the network input, a new unknown case appears which the network is unable to recognise, it means that it is different from all the classes known previously. The SOFM network taught in this way can serve as a detector signalling the appearance of a widely understood novelty. Such a network can also look for similarities between the known data and the noisy data. In this way, it is able to identify fragments of images presenting photographs of orchard pests, for example. The resulting model of the Kohonen neural turns to be effective without reference classifier. The average classification error SOFM network during its operation was 0.05532 for the learning set and 0.0762 for the validtion set.