Wyniki wyszukiwania - BazTech

1

Stripe segmentation of oceanic internal waves in SAR images based on Gabor transform and K-means clustering

Qi Kai-Tuo, Zhang Hong-Sheng, Zheng Ying-Gang, Zhang Yu, Ding Long-Yu

Oceanologia

|

2023

|

Vol. 65 (4)

548--555

EN

Oceanic internal waves are an active ocean phenomenon that can be observed, and their relevant characteristics can be acquired using synthetic aperture radar (SAR). The locations of oceanic internal waves must be determined first to obtain the important parameters of oceanic internal waves from SAR images. An oceanic internal wave segmentation method with integrated light and dark stripes was described in this study. To extract the SAR image characteristics of oceanic internal waves, the Gabor transform was initially used, and then the K-means clustering algorithm was used to separate the light (dark) stripes of oceanic internal waves from the background in the SAR images. The regions of the dark (light) stripes were automatically determined based on the differences between the three classes, that is, the dark stripes, light stripes, and background area. Finally, the locations of the dark (light) stripes were determined by shifting a given distance along the normal direction of the long side with the minimum bounding rectangle of the light (dark) stripes. The best segmentation results were obtained based on the intersection over the union of the images, and the accuracy of segmentation was verified. Furthermore, the effectiveness and practicability of the proposed method in the light and dark stripe segmentation of SAR images of oceanic internal waves were illustrated. The proposed method prepares the foundation for future inversion studies of oceanic internal waves.

2

A combination K-means clustering and 2-opt algorithm for solving the two echelon e-commerce logistic distribution

Zuhanda Muhammad Khahfi, Suwilo Saib, Sitompul Opim, Mardiningsih Mardiningsih

LogForum

|

2022

|

Vol. 18, no. 2

213--225

EN

Background: The rise of e-commerce in the community makes competition between logistics companies increasingly tight. Every e-commerce application offers the convenience and choices needed by the community. The Two-Echelon Vehicle Routing Problem (2E-VRP) model has been widely developed in recent years. 2E-VRP makes it possible for customers to combine shipments from several different stores due to satellites in their distribution stream. The aim of this paper is to optimize a two-echelon logistics distribution network for package delivery on e-commerce platforms, where vans operate in the first echelon and motorcycles operate in the second echelon. The problem is formulated as 2E-VRP, where total travel costs and fuel consumption are minimized. This optimization is based on determining the flow in each echelon and choosing the optimal routing solution for vans and motorcycles. Methods: This paper proposes a combination of the K-means Clustering Algorithm and the 2-opt Algorithm to solve the optimization problem. Many previous studies have used the K-means algorithm to help streamline the search for solutions. In the solution series, clustering is carried out between the satellite and the customer in the first echelon using the K-means algorithm. To determine the optimal k-cluster, we analyzed it using the silhouette, gap statistic, and elbow methods. Furthermore, the routing at each echelon is solved by the 2-opt heuristic method. At the end of the article, we present testing of several instances with the different number of clusters. The study results indicate an influence on the determination of the number of clusters in minimizing the objective function. Results: This paper looks at 100 customers, 10 satellites, and 1 depot. By working in two stages, the first stage is the resolution of satellite and customer problems, and the second stage is the resolution of problems between the satellite and the depots. We compare distance and cost solutions with a different number of k-clusters. From the test results, the number of k-clusters shows an effect of number and distance on the solution. Conclusions: In the 2E-VRP model, determining the location of the cluster between the satellite and the customer is very important in preparing the delivery schedule in logistics distribution within the city. The benefit is that the vehicle can divide the destination according to the location characteristics of the satellite and the customer, although setting the how many clusters do not guarantee obtaining the optimal distance. And the test results also show that the more satellites there are, the higher the shipping costs. For further research, we will try to complete the model with the metaheuristic genetic algorithm method and compare it with the 2-opt heuristic method.

3

Radial basis function neural network and salp swarm algorithm for paddy leaf diseases classification in Thanjavur, Tamilnadu geographical region

Thirugnanasambandam Gayathri Devi, Rajkumar Ganesan, Srinivasan Anandan, Sudha Selvarajan

Acta Geophysica

|

2022

|

Vol. 70, no 6

2917--2932

EN

Farming is an essential sustenance for the progressive population. The development of our country depends on the farmers. Plants endure by many diseases due to environmental factors. So, the farmers need to detect plant diseases at an early stage for appreciable yield. In the beginning, the observing and examining plant disease are examined physically by the expertise in the farming field, which requires a considerable measure of work/ and requires over the top handling time. Now, machine learning concepts eliminate conventional protruding and time-consuming techniques. This paper focuses on a novel method for detecting and identifying paddy leaf diseases at the early stages in Thanjavur region using radial basis function neural network (RBFNN) classifier. Further, it is optimized with salp swarm algorithm (SSA) technique. The proposed method utilizes the data from the TNAU agritech portal, IRRI knowledge bank, UCI machine learning repository databases, which have healthy and diseased images. This work illustrates four categories (Bacterial Blast, Bacterial Blight, Leaf Tungro and Brown Spot) of infected paddy images along with the normal set of images. Initially the preprocessing is performed for the acquired images then K-means segmentation algorithm segregates the image. Gray level co-occurrence matrix extracts the Texture features from the segmented image and the RBFNN classifier performs the disease classification and improves the detection accuracy by optimizing the data using SSA. The investigational results of the proposed methodology exhibit the performance in terms of accuracy of disease detection is 98.47%. However, radial basis function neural network (RBFNN) achieves the diseases detection accuracy of 97.85% and support-vector machine (SVM) classifier achieves a disease detection accuracy of 97.07%. This paper proposes a method of paddy leaf disease recognition and classification using RBFNN and salp swarm algorithm. It also suggests and identifies an image analysis by framing a set of conditions for disease affected plants. The results show that the most satisfactory outcome can be gained to verify the yield of proposed methods with least effort.

4

A novel melanoma detection model: adapted K-means clustering-based segmentation process

Sukanya S. T., Jerine

Bio-Algorithms and Med-Systems

|

2021

|

Vol. 17, no. 2

103--118

EN

Objectives: The main intention of this paper is to propose a new Improved K-means clustering algorithm, by optimally tuning the centroids. Methods: This paper introduces a new melanoma detection model that includes three major phase’s viz. segmentation, feature extraction and detection. For segmentation, this paper introduces a new Improved K-means clustering algorithm, where the initial centroids are optimally tuned by a new algorithm termed Lion Algorithm with New Mating Process (LANM), which is an improved version of standard LA. Moreover, the optimal selection is based on the consideration of multi-objective including intensity diverse centroid, spatial map, and frequency of occurrence, respectively. The subsequent phase is feature extraction, where the proposed Local Vector Pattern (LVP) and Grey-Level Co-Occurrence Matrix (GLCM)-based features are extracted. Further, these extracted features are fed as input to Deep Convolution Neural Network (DCNN) for melanoma detection. Results: Finally, the performance of the proposed model is evaluated over other conventional models by determining both the positive as well as negative measures. From the analysis, it is observed that for the normal skin image, the accuracy of the presented work is 0.86379, which is 47.83% and 0.245% better than the traditional works like Conventional K-means and PA-MSA, respectively. Conclusions: From the overall analysis it can be observed that the proposed model is more robust in melanoma prediction, when compared over the state-of-art models.

5

Homogeneous regionalization via L-moments for Mumbai City, India

Parchure Amit Sharad, Gedam Shirish Kumar

Meteorology Hydrology and Water Management. Research and Operational Applications

|

2019

|

Vol. 7, Iss. 2

73--83

EN

This study identified homogeneous rainfall regions using a combination of cluster analysis and the L-moments approach. The L-moments of heavy rainfall events of various durations (0.25, 1, 6, 12, 24, 48, 72, 96, and 120 h) were analysed using seasonal (June-September) rainfall measurements at 47 meteorological stations over the period 2006- 2016. In the primary phase of this study, the homogeneity of Mumbai as a single region was examined by statistical testing (based on L-moment ratios and variations of the L-moments). The K-means clustering approach was applied to the site characteristics to identify candidate regions. Based on the most appropriate distribution, these regions were subsequently tested using at-site statistics to form the final homogeneous regions. For durations above 1h, the regionalisation procedure delineated six clusters of similarly behaved rain gauges, where each cluster represented one separate class of variables for the rain gauges. However, for durations below 1h, the regionalisation procedure was not efficient in the sense of identifying homogeneous regions for rainfall. Furthermore, the final clusters confirmed that the spatial variation of rainfall was related to the complex topography, which comprised flatlands (below or at mean sea level), urban areas with high rise buildings, and mountainous and hilly areas.

6

Dynamic Measurement of Foam-Sized Yarn Properties from Yarn Sequence Images

Zhu B., Li Z., Cao X., Liu J., Gao W.

Autex Research Journal

|

2018

|

Vol. 18, no. 3

314--322

EN

Unlike the normal sizing method, the foam sizing had been proven to be a low-add-on technology. To investigate the effect of foam sizing, film thickness, sized-yarn evenness, and size penetration rate were necessary to evaluate the performances of foam-sized yarns. However, the conventional image analysis of sized-yarn cross sections primarily relied on artificial testing with a low efficiency. This paper proposed a novel dynamic method to measure the sized-yarn properties including film thickness, sized-yarn evenness, and size penetration rate based on yarn sequence images captured from a moving yarn. A method of dynamic threshold module was adopted to obtain threshold for segmenting yarns in the sequence images. K-means clustering algorithm was applied to segment pixels of the images into yarn and background. To further remove burrs and noise in the images, two judgment templates were carried out to extract the information of yarn core. The film thickness, sized-yarn evenness, and size penetration rate were measured based on the yarn core of each frame in sequence images. In order to compare with the experimental results of the dynamic method, the yarn properties of the same samples were tested by static and artificial testing. Results revealed that the proposed method could efficiently and accurately detect the film thickness, sized-yarn evenness, and size penetration rate.