CONCEPT PAPER | doi:10.20944/preprints202007.0002.v1
Subject: Engineering, Electrical & Electronic Engineering Keywords: Forensic; Ink Mismatch; Clustering; K-means algorithm; Elbow; Silhouette
Online: 1 July 2020 (09:01:50 CEST)
In document forensic, Ink mismatch relays very important information about forgeries in this way we can find out the authenticity of documents. Finding out and distinguishing these unique inks from the multispectral document is very challenging task. In this paper we proposed the method to identify the inks using clustering. We used K-Mean clustering instead of widely known Fuzzy C-Means Clustering (FCM) and successfully identity the number of inks. For the purpose of optimizing and improving our results we used two optimization techniques such as Elbow and silhouette optimization techniques.
Subject: Engineering, Automotive Engineering Keywords: light pollution; silhouette; shutter; cultural heritage lighting; optimized exterior lighting
Online: 23 June 2021 (13:08:48 CEST)
Improper illumination of cultural heritage buildings and monuments can be a source of enormous light pollution that can have a major impact on the overall environment of the illuminated area. Such lighting can be the result of carelessness or a wrong/badly designed lighting system. This paper presents one of the methodologies that can significantly reduce light pollution, especially the spilt light out of the façade. The methodology is based on using luminaires with specially made shutters with the appropriate silhouette of the object. The shutter is planned with the help of an object photo and sizes measured at the location. The methodology was tested during the renovation of the lighting system of different churches in Slovenia and is described in the example of the Church of St. Thomas near Ptuj, Slovenia. The results showed that the methodology is effective and can significantly reduce light pollution that occurs when such buildings are incorrectly lit.
ARTICLE | doi:10.20944/preprints202212.0082.v1
Subject: Mathematics & Computer Science, Computational Mathematics Keywords: K-means clustering algorithm; Elbow method; Silhouette technique; Kneedle Algorithm; Image Segmentation; Conventional Neural Network.
Online: 6 December 2022 (01:30:03 CET)
The agricultural sector in Palestine has a significant role in its economy. However, the production of this sector is affected by different kinds of plant diseases, specifically leaf diseases. Automatic agricultural leaf disease detection is essential for the early diagnosis and controls the overall health of fields. Image segmentation techniques, clustering, and deep learning are often used to detect diseased leaves. This study proposes a novel hybrid approach based on image classification. The hybrid approach combines the k-means clustering algorithm with Convolutional Neural Network (CNN), where k-means is used to detect the leaf’s infected area, then CNN is used for specifying the disease. We used the PlantVillage dataset for experimental verification as it contains several crops with different kinds of challenging diseases. We also examined the selection of optimal k-value using the Silhouette coefficient, Elbow method, and Kneedle Algorithm. The Silhouette technique was analyzed using three distance metrics; Euclidean, Manhattan, and Cosine. Its scores for the three-distance metrics were low, near-zero, and failed to produce the optimum k value. Besides, the Elbow method was complicated to use in image segmentation in terms of executing and visualizing the k value in its graph plot. Based on verification results, the Kneedle Algorithm produced better results in the consistency of choosing the optimal k value and showed superiority over other approaches. Therefore, the processed images were segmented with the k-means clustering algorithm with a Kneedle algorithm-based k value. Finally, a Convolutional Neural Network (CNN) is trained to classify the type of disease based on analyzing and testing leaf images. The hybrid model achieved high accuracy of 93.79% in disease identification, confirming the proposed model’s robustness.
ARTICLE | doi:10.20944/preprints202008.0254.v1
Subject: Mathematics & Computer Science, Information Technology & Data Management Keywords: feature selection; k-means; silhouette measure; clustering; big data; fault classification; sensor data; time-series data
Online: 11 August 2020 (06:26:43 CEST)
Feature selection is a crucial step to overcome the curse of dimensionality problem in data mining. This work proposes Recursive k-means Silhouette Elimination (RkSE) as a new unsupervised feature selection algorithm to reduce dimensionality in univariate and multivariate time-series datasets. Where k-means clustering is applied recursively to select the cluster representative features, following a unique application of silhouette measure for each cluster and a user-defined threshold as the feature selection or elimination criteria. The proposed method is evaluated on a hydraulic test rig, multi sensor readings in two different fashions: (1) Reduce the dimensionality in a multivariate classification problem using various classifiers of different functionalities. (2) Classification of univariate data in a sliding window scenario, where RkSE is used as a window compression method, to reduce the window dimensionality by selecting the best time points in a sliding window. Moreover, the results are validated using 10-fold cross validation technique. As well as, compared to the results when the classification is pulled directly with no feature selection applied. Additionally, a new taxonomy for k-means based feature selection methods is proposed. The experimental results and observations in the two comprehensive experiments demonstrated in this work reveal the capabilities and accuracy of the proposed method.