Olivia Mendoza

An improved method for edge detection based on interval type-2 fuzzy logic

In this paper, a method for edge detection in digital images based on the morphological gradient and fuzzy logic is described. A basic method for edge detection was improved using fuzzy logic. An advantage of the improved method is that there is no need of applying filtering to the image. The simulation results were obtained with a type-1 fuzzy inference system (T1FIS) and with an interval type-2 fuzzy inference system (IT2FIS) for improving the edge detection method. We show that the images obtained with fuzzy logic are better than the ones obtained with only the morphological gradient method. In particular the IT2FIS achieved the best results, because of the flexibility to model the uncertainty in the gradient values and the gray ranges for the edge images. In both TIFIS and IT2FIS the membership function parameters were obtained directly from the images; this allows application of the proposed method to images with different gray scales.

Edge-Detection Method for Image Processing Based on Generalized Type-2 Fuzzy Logic

This paper presents an edge-detection method that is based on the morphological gradient technique and generalized type-2 fuzzy logic. The theory of alpha planes is used to implement generalized type-2 fuzzy logic for edge detection. For the defuzzification process, the heights and approximation methods are used. Simulation results with a type-1 fuzzy inference system, an interval type-2 fuzzy inference system, and with a generalized type-2 fuzzy inference system for edge detection are presented. The proposed generalized type-2 fuzzy edge-detection method was tested with benchmark images and synthetic images. We used the merit of Pratt measure to illustrate the advantages of using generalized type-2 fuzzy logic.

A hybrid approach for image recognition combining type-2 fuzzy logic, modular neural networks and the Sugeno integral

In this paper, a hybrid approach for image recognition combining type-2 fuzzy logic, modular neural networks and the Sugeno integral is described. Interval type-2 fuzzy inference systems are used to perform edge detection and to calculate fuzzy densities for the decision process. A type-2 fuzzy system is used for edge detection, which is a pre-processing applied to the training data for better use in the neural networks. Another type-2 fuzzy system calculates the fuzzy densities necessary for the Sugeno integral, which is used to integrate results of the neural network modules. In this case, fuzzy logic is shown to be a good methodology to improve the results of a neural system facilitating the representation of the human perception. A comparative study is also made to verify that the proposed approach is better than existing approaches and improves the performance over type-1 fuzzy logic.

Interval type-2 fuzzy logic and modular neural networks for face recognition applications

In this paper we present a method for response integration in multi-net neural systems using interval type-2 fuzzy logic and fuzzy integrals, with the purpose of improving the performance in the solution of problems with a great volume of information. The method can be generalized for pattern recognition and prediction problems, but in this work we show the implementation and tests of the method applied to the face recognition problem using modular neural networks. In the application we use two interval type-2 fuzzy inference systems (IT2-FIS); the first IT2-FIS was used for feature extraction in the training data, and the second one to estimate the relevance of the modules in the multi-net system. Fuzzy logic is shown to be a tool that can help improve the results of a neural system by facilitating the representation of human perceptions.

A hybrid modular neural network architecture with fuzzy Sugeno integration for time series forecasting

We describe in this paper the application of a modular neural network architecture to the problem of simulating and predicting the dynamic behavior of complex economic time series. We use several neural network models and training algorithms to compare the results and decide at the end, which one is best for this application. We also compare the simulation results with the traditional approach of using a statistical model. In this case, we use real time series of prices of consumer goods to test our models. Real prices of tomato in the U.S. show complex fluctuations in time and are very complicated to predict with traditional statistical approaches. For this reason, we have chosen a neural network approach to simulate and predict the evolution of these prices in the U.S. market.

An Interval Type-2 Fuzzy Neural Network for Chaotic Time Series Prediction with Cross-Validation and Akaike Test

A novel homogeneous integration strategy of an interval type-2 fuzzy inference system (IT2FIS) with Takagi-Sugeno-Kang reasoning (TSK IT2FIS) is presented. This TSK IT2FIS is represented as an adaptive neural network (NN) with hybrid learning (IT2FNN:BP+RLS) in order to automatically generate an interval type-2 fuzzy logic system (TSK IT2FLS). Consequent parameters are updated with recursive least-square (RLS) algorithm; antecedent parameters with back-propagation (BP) algorithm. Mackey-Glass chaotic time series forecasting results are presented ((=17, 30, 100) with different signal noise ratio (SNR). Soundness for uncertainty, adaptability and learning and generalization capabilities is shown using 10-fold Cross Validation, Akaike Information Criteria (AIC) and F-Test.

A New Method for Edge Detection in Image Processing Using Interval Type-2 Fuzzy Logic

Edges detection in digital images is a problem that has been solved by means of the application of different techniques from digital signal processing. Also the combination of some of these techniques with fuzzy inference system (FIS) has been applied. In this work a new FIS type-2 method is implemented for the detection of edges and the results of three different techniques for the same goal are compared.

Type-2 Fuzzy Logic for Improving Training Data and Response Integration in Modular Neural Networks for Image Recognition

The combination of Soft Computing techniques allows the improvement of intelligent systems with different hybrid approaches. In this work we consider two parts of a Modular Neural Network for image recognition, where a Type-2 Fuzzy Inference System (FIS 2) makes a great difference. The first FIS 2 is used for feature extraction in training data, and the second one to find the ideal parameters for the integration method of the modular neural network. Once again Fuzzy Logic is shown to be a tool that can help improve the results of a neural system, when facilitating the representation of the human perception.

Modular Neural Networks and Type-2 Fuzzy Logic for Face Recognition

In this paper we present a method for face recognition combining modular neural networks and two interval type-2 fuzzy inference systems (FIS 2) for face recognition. The first FIS 2 is used for edges detection in the training data, and the second one to find the ideal parameters for the Sugeno integral as a decision operator. Fuzzy logic is shown to be a tool that can help improve the results of a neural system facilitating the representation of the human perception.

Comparison of Fuzzy Edge Detectors Based on the Image Recognition Rate as Performance Index Calculated with Neural Networks

Edge detection is a process usually applied to image sets before the training phase in recognition systems. This preprocessing step helps to extract the most important shapes in an image, ignoring the homogeneous regions and remarking the real objective to classify or recognize. Many traditional and fuzzy edge detectors can be used, but it’s very difficult to demonstrate which one is better before the recognition results. In this work we present an experiment where several edge detectors were used to preprocess the same image sets. Each resultant image set was used as training data for neural network recognition system, and the recognition rates were compared. The goal of this experiment is to find the better edge detector that can be used as training data on a neural network for image recognition.