Piotr Boniecki

Neural identification of selected apple pests

The study was based on neural network modeling methods, including image analysis.Artificial neural networks as a powerful tool to identify pests.The color of pests as the dominant input variable of a neural model.The best classification ability was achieved by MLP topology. The subject of this study was to investigate the possibility of using artificial neural networks as a tool for classification, designed to identify apple orchard pests. The paper presents a classification neural model using optimized learning sets acquired on the basis of the information encoded in the form of digital images of selected pests. This study predominantly deals with the problem of the identification of 6 selected apple pests which are most commonly found in Polish orchards. Neural modeling techniques, including digital image analysis, were used to classify the pests.The qualitative analysis of neural models produced, indicates that multi-layered perceptron (MLP) neural network topology achieve the best classification ability. Representative features, allowing for effective pest identification are 23 visual parameters in the form of 7 selected coefficients of shape and 16 color characteristic of pests. The dominant input variables of a neural model, determining the correct identification of the features, contain information about the color of pests.Our results support the hypothesis that artificial neural networks are an effective tool that supports the process of identification of pests in apple orchards. The resulting neural classifier has been created to assist in the decision-making processes that take place during the production of apples, in the context of protection against pests.

The selected examples of the application of computer image analysis in the assessment of environmental quality

The environmental monitoring (EM) is an essential part of protection of the environment, most of the methods of environmental protection based on visual techniques or physico-chemical and biochemical measurements. The automation of traditional methods proceeds at an accelerating rate, modern laboratories prefer this type of tools to conduct a more comprehensive assessment and online monitoring. The application of computer image analysis methods in biomonitoring brings to this discipline the opportunity to develop innovative tools that allow for more precise sensitive and quantified assessment of monitored processes. The application of techniques based on computer image processing technology will dominate in the future and very comfortable and intuitive tool for researchers in the study of the components of the environment quality. The article presents some methods of automation the acute toxicity bioassay based on the application of computational methods.

Possibilities of neural image analysis implementation in monitoring of microalgae production as a substrate for biogas plant

The paper presents the possibilities of neural image analysis of microalgae content in the large-scale algae production for usage as a biomass. With the growing conflict between the culture produced both for feed and energetic purpose in Europe, the algae production seems to be very efficient way to produce the huge amount of biomass outside of conventional agronomy. However, for stable microalgae production the key point for culture management is the rapid estimation of algae population and assessment of its developmental stage. In traditional way the microalgae content is usually checked by the long microscopic analyses which cannot be used in large-scale industrial cultivation. Moreover, highly specialized personnel is required for algal determinations. So the main aim of this study is to estimate the possibility of usage of automatic image analysis of microalgae content made by artificial neural network. The preliminary results show that the selection of artificial neural network topology for the microalgae identification allowed for the selection and choice of teaching variables obtained by studying the image analysis. The selected neural model on the basis of data from computer image analysis allows to carry out the operations of algae identification and counting. On the basis of the obtained results of preliminary tests it is possible to count the algae on the photos. Additional information on their size and color allows to unlimited categorization.

Neural Image Analysis of Maturity Stage during Composting of Sewage Sludge

The paper presents the experiments of compost images analysis carried out with two types of digital cameras working in daylight and ultraviolet light. The data collected with two cameras were analysed with the usage of neural network model (using part of application Statistica v. 8.0). The results of image analysis were combined also with the results of chemical and physical analysis of composted material in different stage of the composting process.

Identification of malting barley varieties using computer image analysis and artificial neural networks

The project aimed to produce an identification model that allows for automatic recognition of malting barley varieties. The project used computer image analysis and artificial neural networks. The authors based on the analysis of biological material selected set of features describing the physical parameters allowing the identification of varieties. Image analysis of samples of barley digital photographs allowed the extraction of the characteristics of varieties. Obtained characteristics from the images were used as learning data for artificial neural network. Trained a multilayer perceptron network is characterized by the identification abilities at the level of human abilities.

Use of artificial neural networks in the identification and classification of tomatoes

The project aimed to produce a classification model of neural network that would allow automatic evaluate quality of greenhouse tomatoes. The project used computer image analysis and artificial neural networks. Authors based on the analysis of biological material selected set of features that are describing the physical parameters allowing the quality class identification. Image analysis of tomatoes digital photographs samples allowed to choose characteristics features. Obtained characteristics from the images were used as learning data for artificial neural network.

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.

The Identification of Mechanical Damages of Kernels Basis on Neural Image Analysis

The aim of the study was to develop a neural model for the identification of mechanical damage in maize caryopses based on digital photographs. The author has selected a set of features that distinguish between damaged and healthy caryopses. The study has produced an artificial neural network of a multilayer perceptron type whose identification capacity approximates that of a human.

Computer image analysis in the quality in procedure for selected carrot varieties

In our daily lives we often assess our surroundings to classify the situations we encounter. We do so based on the observations we make of our surroundings and information we obtain from other sources, using our knowledge and abilities. While this process is natural to us, if we want to give a similar task to a computer system then we have to take various steps in order to enable our computers to partially emulate the human capacity for observation, learning and making final decisions based on knowledge. As information complexity increases, there is an increasing demand for systems which can recognize and classify the objects presented to them. Recently there has been an increase in interest in application of computer image analysis in various research areas. One of these applications is food quality assessment, which aims to replace traditional instrumental methods. A computer visual system was developed to assess carrot quality, based on a single variety. Characteristic qualities of the variety were chosen to describe a suitable root. In the course of the study, digital photographs of carrot roots were taken, which were used as input data for the assessment performed by a dedicated computer program created as a part of the study.

A stand for the image acquisition of composted material based on the sewage sludge

Composting is one of the best methods for sewage sludge management. The early identification of the young compost stage in composted material is important. The method for determining the degree of maturity of composted material containing sewage sludge will use the selected topologies of artificial neural networks. The learning processes of these networks will be carried out with the use of the information contained in digital images of composted material. It is important that acquisition of these images was carried out under constant lighting and exposure conditions on a suitable acquisition stand. The objectives of presented study were: to develop a stand for image acquisition of composted material, to determine the spectral distribution for used light sources and illuminance distribution for visible light, to determine the parameters for image acquisition of composted material. A suitable stand, consisted of three photographic chambers illuminated with visible light, UV-A light and mixed light, was developed. The spectral distribution of the used light sources and the illuminance distribution for visible light were analyzed and considered satisfactory. Image acquisition parameters, such as focal length, ISO sensitivity, aperture and exposure time, were specified.