Jacky Desachy

Numeric and symbolic data fusion: a soft computing approach to remote sensing images analysis

Abstract An expert system approach for image classification according to expert knowledge about best sites for vegetation classes is described. Uncertainty management is solved by a certainty factor approach. The numerical and symbolic data fusion is viewed as an updating process. The fusion approach is then described. A neural classifier applied to image data is the first source. A set of fuzzy neural networks representing expert knowledge constitutes the second source. A conjunctive combination based on evidence theory is applied. Finally, a possibility theory-based pooling aggregation rule is presented. These three approaches are applied to a vegetation classification problem.

Geographic information system updating using remote sensing images

Abstract In this paper, we propose several methods to update and upgrade GIS using remote sensing images. In the first part, we present a matching method of GIS vectors on SPOT images, allowing the localisation of geographic linear elements to be improved; and in the second part, we present an extraction method of new linear elements using Ikonos images, with a multi-resolution approach.

Evaluation of biochemical sources pertinence in classification of cell's physiological states by evidence theory

For analysis and modelling of the biotechnological process we must look deeper to the biological systems. Everybody knows that cell metabolism and the resulting kinetic is a complex process which could not be modelling completely by a non-linear differential system. The goal of all modelling is either the biocontrol or finding the physiological states. We want to detect the physiological states using a small number of measured signals. We present in this paper the analyses of biochemical parameters using the evidence theory. The evidence theory is also used to characterize the pertinence of the parameters. This pertinence is based on the notion of conflict. We show that our measure of conflict based on a distance provides more coherent results as the classical methods. Based on the analysis of microbiological process a method has been implemented for fermentation real-time analysis.

Detection and characterization of physiological states in bioprocesses based on Hölder exponent

Today, the pace of progress in fermentation is fast and furious, particularly since the advent of genetic engineering and the recent advances in computer sciences and process control. The high cost associated with many fermentation processes makes optimization of bioreactor performance trough command control very desirable. Clearly, control of fermentation is recognized as a vital component in the operation and successful production of many industries. Todays advances in measurement, data acquisition and handling technologies provide a wealth of new data which can be used to improve existing models. In this article we propose a method of physiological state identification based on segmentation of bioreactor sensors signals. The underlying of this method is based on the detection of signals singularities by the Maximum of Modulus of Wavelets Transform and their characterization by Holder exponent evaluation. The physiological states identification is based on the correlation product between biochemical signals. The efficiency of the method has been tested in a fed-batch fermentation having the goal to increase the biomass production.

Combination of remote sensing and geocoded data for satellite image interpretation based on neural networks

Presents an image interpretation system for automatic cartography using neural networks. In this frame, the results of two neural networks are combined for producing a final map. The first network deals with radiometric satellite data and the second one with geocoded data (such as digital elevation models, road maps, and hydrographic network). In both cases, the authors propose new methods for, on the one hand, the maximum likelihood method, and on the other hand, expert knowledge based classification using production rules with certainty factors. The authors also introduce future works, that is to say rule extraction.<<ETX>>

Extraction of geographic features using multioperator fusion

Automatic analysis of remote sensing images faces different problems: context diversity, complexity of information. To simplify identification and to limit the search space, we use extra data and knowledge to help the scene understanding. Diversity and imprecision of information sources generate new problems. The fuzzy logic theory is used to solve the problem of imprecision. Many extraction algorithms are used to provide a more reliable result. Extraction may be performed either globally on the whole image or locally using information of data bases. Each extractor produces a map of certainty factors for a given type of geographic features according to their characteristics: radiometry, color, linear, etc. Maps contain wrong detections due to imperfections of the detectors or non- completeness of generic models. So, we generate a new map using fusion to have a best credibility used to compute a dynamic programming. It finds an optimal path even if the linear feature is partially occluded. But the path is generally erratic due to noise. Then a snake-like technique smooth the path to clean the erratic parts and to tune the level of detail required to represent the geographic features on a map of a given scale. The result is used to update data bases.

Pattern Recognition by Splitting Images Into Trees of Fuzzy Regions

In this paper, a new method of pattern recognition based on images splitting into a set of trees composed of fuzzy regions is presented. First, either a gradient inverse function is applied on the raster image to define the fuzzy regions supports, or we manage with the basic grey level image if regions are easily topologically separable. Then, topologic features are computed on these sets. Therefore, a tree description of the image, which consists of fuzzy regions with associated topological features, is obtained. A set of sample trees is achieved from the application of the fuzzy segmentation algorithm on characteristic objects small images. Then a tree isomorphism is defined to recognize a particular object in an image. At last, a new tree compression method is introduced in order to decrease the complexity when we have to manage with a large set of samples.

Estimation of relevance and fusion of data sources using belief function theory: application to bioprocess

In this paper, we present an application of the belief function theory for the classification of physiological states in a bioprocess. It also takes account of the relevance of the data sources. The notion of conflict is used to evaluate the relevance of each data source. Another measure of conflict, based on a distance, is also used, and provides globally, better results than the classical notion of conflict used in the Dempster rule of fusion. Experimental results are presented for a bioprocess and show that, with the use of relevance, the results of classification are better.

Automatic tracking of linear features on SPOT images using dynamic programming

Detection of geographic elements on images is important in the perspective of adding new elements in geographic databases which are sometimes old and so, some elements are not represented. Our goal is to look for linear features like roads, rivers or railways on SPOT images with a resolution of 10 meters. Several methods allow this detection to be realized and may be classified in three categories: (1) Detection operators: the best known is the DUDA Road Operator which determine the belonging degree of a pixel to a linear feature from several 5 X 5 filters. Results are often unsatisfactory. It exists too the Infinite Size Exponential Filter (ISEF), which is a derivative filter and allows edge, valley or roof profile to be found on the image. It can be utilized as an additional information for others methods. (2) Structural tracking: from a starting point, an analysis in several directions is performed to determine the best next point (features may be: homogeneity of radiometry, contrast with environment, ...). From this new point and with an updated direction, the process goes on. Difficulty of these methods is the consideration of occlusions (bridges, tunnels, dense vegetation, ...). (3) Dynamic programming: F* algorithm and snakes are the best known. They allow a path with a minimal cost to be found in a search window. Occlusions are not a problem but two points or more near the searched linear feature must be known to define the window. The method described below is a mixture of structural tracking and dynamic programming (F* algorithm).

Information Sources Fusion Approach in Forest Stand Classification

The classification of remote-sensing images based on multiple information sources offers a consistent method for the automatic cartography of forest stands. However, fusion models reveal problems of combinatorial explosion due to the calculation of the assignment functions. This article proposes an information-fusion approach that responds to the need for updating the forest inventory, based on belief theory. It illustrates a solution that overcomes the problem of combinatorial explosion that arises with the evaluation of evidence-mass functions which are used as the frame of discernment events. This solution is based on a refinement of the frame of discernment based on the determination of all focal elements (singleton or composite hypothesis of non null masses). Thus, the combination of information source masses would involve only the focal elements masses. In the approach proposed here, the notions of fuzzy logic and possibility theory have been used for the calculation of masses and combinations between classes as an intermediary phase in arriving at belief functions. The result of the application of our fusion approach revealed a significant improvement in optimizing the calculation of mass evidence functions and thus achieving a satisfactory classification.