Mapping adaptive fuzzy Kohonen clustering network onto distributed image processing system

Mining event logs for knowledge discovery based on adaptive efficient fuzzy Kohonen clustering network

Software components analysis is a critical technique for software development and maintenance. Clustering techniques have been widely used in grouping related software components. However, software is complex, but clustering techniques used in software engineering typically adopt only one metric to measure the similarity of components. This paper proposes an adaptive fuzzy clustering technique based on possibilistic clustering algorithms to address the issue of single metric. The proposed technique collaboratively considers distance, density, and the trend of density change of component instances in the membership degree calculation. The post clustering separation of clustered components based on the predefined thresholds and regrouping of the separated component points result in higher cohesive clustering. The proposed algorithm has been evaluated via experiments using a network protocol RSVP-TE system. The comparison of Hierarchical, Self-organizing map (SOM), and fuzzy c-means (FCM) against adaptive fuzzy clustering proposed in this paper indicates that the adaptive fuzzy clustering group software component instances into more cohesive clusters while it is also insensitive to parameter settings.

A fuzzy self-tuning controller based on fuzzy clustering algorithmis proposed in the paper. First of all, the adaptive fuzzy clustering algorithm is presented to confirm the structure of fuzzy model. The Kalman filtering method is used to confirm parameters of fuzzy model. On the basis of the proposed method, the parameters of the controller can be automatically adjusted. To demonstrate the performance of the proposed method, simulation on the nonlinear system is performed. The results show that this method provides effective and accurate control.

In reverse engineering one often needs to generate geometrical models of old mechanical parts. Manual redrafting of old products using computer aided design (CAD) software is a laborious procedure. To overcome this problem, an automated procedure is proposed. The proposed approach is based on using clustering to segment the range image of the object and then constructing a solid model using an automated procedure. To demonstrate the feasibility of the proposed method, the model generation for convex planar faceted objects is considered. A generalized version of the adaptive fuzzy c-elliptotype clustering algorithm suitable for higher dimensional data is used to identify the planar facets in noisy range images when the number of facets is unknown.

Statistical as well as adaptive clustering approaches are being currently used for both segmentation and vector quantization of medical images. However, a comparative evaluation of both approaches has rarely been done to identify the efficacy of such approaches to specific applications, for example, image segmentation and vector quantization. The rate distortion functions of three clustering algorithms, namely, the statistical based deterministic annealing, the adaptive fuzzy leader clustering algorithm, and LBG, have been computed for vector quantization using multi-scale vectors in the wavelet domain. Such comparative evaluation serves as a guide for proper selection of clustering algorithms for global codebook generation in vector quantization and for image segmentation.

The silicon content of molten iron is a key index to reveal the thermal state in the blast furnace (BF). To improve the quality of molten iron, it is important to obtain silicon content online accurately. However, owing to the harsh environment in BF hearth, it is difficult to realize the online measurement of the silicon content of molten iron. To handle this problem, a novel soft sensoring method based on adaptive stacked polymorphic model is proposed. First, an adaptive uncertain fuzzy clustering algorithm (AUFC) is utilized to obtain the distribution characteristics and reduce the complexity of the process data, thereby the training samples are divided into several sub-classes. Second, the adaptive weight echo state network (AW-ESN) and time difference AW-ESN (TD-AWESN) are proposed to establish polymorphic models. Inspired by the stacking strategy, sub-models based on stacked polymorphic model are designed, which can effectively solve the time-varying and the strong nonlinear problems in the BF ironmaking process. Third, to avoid model degradation caused by the changes of working conditions, an ensemble strategy is adopted to combine the predicted results of multiple sub-models. The experimental results in an industrial BF show that the soft sensor based on an adaptive polymorphic model can reach the required prediction accuracy and provide reliable silicon content.

The designs of solar energy systems depend mainly on the solar radiation that reaches the earth's surface, as it is difficult to determine the amount of solar radiation precisely due to several climatic, geographical and temporal factors. Therefore, forecasting of solar radiation is necessary before using solar energy systems. In this paper, the researchers present an accuracy MAMDANI fuzzy inference system for solar radiation prediction with meteorological data. This system is based with a two-stage method for Fuzzy Rules Learning through Clustering (FRLC). In the first stage, the subtractive clustering is used to extract the fuzzy rules, the second stage is a linguistic approximation and a refinement of the learned solutions with linguistic hedges. FRLC is compared to multilayer feed-forward neural network and support vector regression. The results of the experiments show the efficacy of linguistic fuzzy rules in the forecasting of solar radiation. In parallel with the prediction, the model provides a good balance between interpretability and accuracy.

Clustering by fast search and find of density peaks (CFSFDP) is proposed to cluster the data by finding of density peaks. CFSFDP is based on two assumptions that: a cluster center is a high dense data point as compared to its surrounding neighbors, and it lies at a large distance from other cluster centers. Based on these assumptions, CFSFDP supports a heuristic approach, known as decision graph to manually select cluster centers. Manual selection of cluster centers is a big limitation of CFSFDP in intelligent data analysis. In this paper, we proposed a fuzzy-CFSFDP method for adaptively selecting the cluster centers, effectively. It uses the fuzzy rules, based on aforementioned assumption for the selection of cluster centers. We performed a number of experiments on nine synthetic clustering datasets and compared the resulting clusters with the state-of-the-art methods. Clustering results and the comparisons of synthetic data validate the robustness and effectiveness of proposed fuzzy-CFSFDP method.

Wireless Sensor Networks are vital for applications like environmental monitoring, smart cities, and industrial Internet of Things, yet their efficiency is often limited by sensor node energy constraints. To address this, we propose a novel hybrid routing algorithm namely, Fuzzy C-Means - Fuzzy-Based Spotted Hyena Optimization - Quantum Annealing, which enhances energy efficiency and network longevity. The three-phase approach begins with adaptive fuzzy clustering, allowing nodes to belong to multiple clusters for improved load balancing. Next, the Fuzzy-Based Spotted Hyena Optimization algorithm selects initial routes using parameters such as residual energy, distance to the base station, and link quality. Finally, Quantum Annealing refines routing paths to ensure global optimization and avoid local optima. This hybrid method adapts dynamically to changing network conditions and energy levels, delivering superior scalability and performance. Experimental results show a 30% reduction in total energy consumption and a 25% increase in effective data transmission. The algorithm achieves a packet delivery ratio of 98.87%, throughput of 65 kbps, and latency of 115 ms, outperforming conventional techniques. This work contributes a robust and adaptive routing solution, supporting sustainable and efficient WSN deployments in dynamic environments.

An improved detection of blind image forgery using hybrid deep belief network and adaptive fuzzy clustering

We compared the performance of two recently developed vector quantization algorithms using different optimization criteria for clustering, namely, the adaptive fuzzy leader clustering, a neuro-fuzzy algorithm, and the deterministic annealing, another unsupervised clustering algorithm based on probabilistic and statistical physics frameworks, with the rate distortion criterion as a performance measure. Such a comparison is useful for evaluating the efficiency of clustering algorithms for the purpose of image vector quantization instead of the conventional misclassification evaluation. This method is extended from the analysis of image coding in a spatial domain to sample vectors in the wavelet domain with predictable distribution. These sample vectors possess a multidimensional generalized Gaussian distribution through the new multi-scale feature extraction method. Our preliminary results show much improvement on the reconstructed image quality over JPEG.

SummaryThe advent of emergent system on chip and multiprocessor system on chip opens a new era on the small mobile devices (smartphones, tablets, etc.) in terms of computing capabilities and applications to be addressed. Given the ability of these devices to interact with the real world through the camera, the development of efficient algorithms related to image processing and computer vision is mandatory. The particle filter algorithm is an algorithm frequently used in image and video processing; it constitutes the baseline algorithm in many applications: feature tracking, facial recognition, tracking of vehicles in traffic, video compression and so on. We propose a parallel implementation for the particle filter algorithm oriented to mobile Android devices. Three different versions of this algorithm are presented: a Java sequential implementation and two Renderscript parallel versions, an ad hoc implementation and a parallel implementation generated automatically with Paralldroid. The results obtained by the parallel versions over different Android platforms present high accuracy with a high processing rate of frame per second and a high speedup at the same time. Copyright © 2015 John Wiley & Sons, Ltd.

A Scalable Infrastructure for Online Performance Analysis on CFD Application

In the current era of large scale distributed computing, users voluntarily donate computation power of their idle computers or desktop machines. However, there is an urgent need of a utility-based distributed problem solving environment to attract more users. Related to applying the utility concept, we are working on design and implementation of a distributed image processing system that uses: 1 contract net protocol for task assignment; 2 concept of utility-based distributed computing; 3 task graph concept for application design. In this paper, we present our results of the design aspects of the system. We propose the design of contract net protocol-based distributed image processing system and analyse characteristics of the proposed system. We evaluate and select available options for implementation of the system. The proposed system allows transparent execution of task graph-based applications in a distributed environment.

Parallelization is the key method to speedup advanced driving assistance systems (ADAS) and autonomous vehicle applications especially using emerging multicore and manycore processors. Localization and map matching are among the fundamental parts of such applications. This paper presents a parallel implementation and performance analysis of particle filter based vehicle localization and map-matching algorithm both on a multicore processor using Open Multi-Processing (OpenMP) and on a graphics processing unit (GPU) using the Compute Unified Device Architecture (CUDA). In the proposed implementation, GPS and odometer data is fused with digital map information. Tests were performed on real data captured in vehicle environment comprising various speed and road conditions. Test results show that speedups up to 75 times can be achieved on parallel GPU implementation over sequential counterpart. As this speedup complies with real-time performance requirements, high computational cost of using map topology information with large number of particles in particle filter implementation is proven to be handled by emerging technologies. The effect of number of particles on error rate of localization and map matching is also investigated.