No abstract

Removing Mixed Gaussian and Impulse Noise (MGIN) is considered to be one of the most essential topics in the domain of image restoration, and it is much more challenging than to remove pure Gaussian or impulse noise separately. Therefore, relatively fewer works have been published in this area. This paper proposes a new integrated approach for MGIN removal that is based on a Non-Singleton Interval Type-2 (NS-IT2) Fuzzy Logic System (FLS), and explains how to design such a NS-IT2 FLS using a Quantum-behaved Particle Swarm Optimization (QPSO) algorithm. Then the paper goes on to introduce two supplementary components, a Block-Matching 3-Dimensional Discrete Cosine Transformation (BM3D DCT) filter and a contrast scaling filter, which augment the overall performance of the NS-IT2 FLS. Finally, the paper shows that this proposed approach indeed provides both quantitatively and visually much better results compared to other often-used non-fuzzy techniques as well as its Type-1 (T1) and singleton IT2 (S-IT2) counterparts.

Removing Mixed Gaussian and Impulse Noise (MGIN) is considered to be very important in the domain of image restoration, but it is a somewhat more challenging topic than removing pure Gaussian or impulse noise. Therefore, relatively fewer works have been published in this area. This paper proposes a Non-Singleton Interval Type-2 (IT2) Fuzzy Logic System (FLS) for MGIN removal, explains how it can be designed based on a Quantum-behaved Particle Swarm Optimization algorithm, and shows that it provides both quantitatively and visually much better results compared to other often-used non-fuzzy techniques as well as its Type-1 and singleton IT2 counterparts.

No abstract

No abstract

The Perceptual Computer (Per-C) is an application of Zadeh's theory of computing with words (CWW) in the field of assisting people to make subjective judgments. This paper describes an application of the Per-C to evaluate Learning Outcomes (LOs) in an outcome-based education (OBE) system. The evaluation problem in this paper is viewed as a hierarchical decision making problem. In the first level of the hierarchy, the crisp score of each assessment is mapped into an interval type-2 fuzzy set (IT2 FS) for each student. The assessments are then aggregated to obtain the performance of each LO. The linguistic weighted power mean (LWPM) is used as the CWW engine, so that some mandatory requirements can be implemented. Finally, by aggregating the performance of all LOs, the final performance is obtained. We also present an example of our proposed Per-C using a fictitious course.

The California's Monterey formation is thought to hold significant hydrocarbon potential and is looked upon as a long term opportunity for development. Given its unique depositional environment, digenesis and very low permeability the hydrocarbon bearing formations have been referred as unconventional reservoirs. Since early 1900's the San Joaquin Valley's Monterey formation has been targeted and produced by several determined and ambitious operators. Still today, these operating companies are trying to develop an understanding of how to best develop and economically unlock this potential to further exploit the resource. Different stimulation techniques, such as hydraulic fracturing and acid treatments, have been used in an attempt to unlock the resource. It was observed that, while hydraulic fracturing was to some extent effective in several fields, other reservoirs have been found to respond very well to acidizing treatments. The wealth of acid treatment data and corresponding production response from the Monterey formations present a great analysis opportunity to identify best practices and optimize well completions to maximize production. This paper presents a study undertaken to analyze the acid stimulation treatments in one of the structures of the Monterey formation and consists of two parts. The first part covers the descriptive analysis and involves the use of supervised and unsupervised clustering techniques to evaluate the relationship between acid treatments and production data. Its objective is to understand the success of the treatments and to identify acid job best practices. In the second part, using only the available information and cluster knowledge extracted, an attempt was made to develop a predictive tool to forecast the performance of new/infill wells and re-stimulation treatments. The study revealed a relatively good understanding of the factors affecting the production response of individual wells and also the variations observed in the different parts of the field. Shortly after the completion of the study, the operating company executed similar activities to what the study found, unbeknownst to the authors. The positive results achieved were not only a validation of the recommendations but also demonstrated the business value created by such analyses.

No abstract