Unsupervised classification methods in food sciences: discussion and outlook

2008 ◽  
Vol 88 (7) ◽  
pp. 1115-1127 ◽  
Author(s):  
Marcin Kozak ◽  
Christine H Scaman
Keyword(s):  
Unsupervised Classification ◽  
Classification Methods

scholarly journals A Novel Unsupervised Classification Method for Sandy Land Using Fully Polarimetric SAR Data

2021 ◽  
Vol 13 (3) ◽  
pp. 355
Author(s):  
Weixian Tan ◽  
Borong Sun ◽  
Chenyu Xiao ◽  
Pingping Huang ◽  
Wei Xu ◽  
...  
Keyword(s):  
Spectral Clustering ◽  
Large Scale ◽  
Clustering Algorithm ◽  
Feature Vector ◽  
Unsupervised Classification ◽  
Classification Method ◽  
Sandy Land ◽  
Classification Methods ◽  
The Many ◽  
Representative Points

Classification based on polarimetric synthetic aperture radar (PolSAR) images is an emerging technology, and recent years have seen the introduction of various classification methods that have been proven to be effective to identify typical features of many terrain types. Among the many regions of the study, the Hunshandake Sandy Land in Inner Mongolia, China stands out for its vast area of sandy land, variety of ground objects, and intricate structure, with more irregular characteristics than conventional land cover. Accounting for the particular surface features of the Hunshandake Sandy Land, an unsupervised classification method based on new decomposition and large-scale spectral clustering with superpixels (ND-LSC) is proposed in this study. Firstly, the polarization scattering parameters are extracted through a new decomposition, rather than other decomposition approaches, which gives rise to more accurate feature vector estimate. Secondly, a large-scale spectral clustering is applied as appropriate to meet the massive land and complex terrain. More specifically, this involves a beginning sub-step of superpixels generation via the Adaptive Simple Linear Iterative Clustering (ASLIC) algorithm when the feature vector combined with the spatial coordinate information are employed as input, and subsequently a sub-step of representative points selection as well as bipartite graph formation, followed by the spectral clustering algorithm to complete the classification task. Finally, testing and analysis are conducted on the RADARSAT-2 fully PolSAR dataset acquired over the Hunshandake Sandy Land in 2016. Both qualitative and quantitative experiments compared with several classification methods are conducted to show that proposed method can significantly improve performance on classification.

scholarly journals Identification of Heracleum sosnowskyi-Invaded Land Using Earth Remote Sensing Data

2020 ◽  
Vol 12 (3) ◽  
pp. 759
Author(s):  
Jūratė Sužiedelytė Visockienė ◽  
Eglė Tumelienė ◽  
Vida Maliene
Keyword(s):  
Remote Sensing ◽  
Invasive Plant ◽  
Remote Sensing Data ◽  
Unsupervised Classification ◽  
Economic Losses ◽  
Classification Methods ◽  
Earth Remote Sensing ◽  
Heracleum Sosnowskyi

H. sosnowskyi (Heracleum sosnowskyi) is a plant that is widespread both in Lithuania and other countries and causes abundant problems. The damage caused by the population of the plant is many-sided: it menaces the biodiversity of the land, poses risk to human health, and causes considerable economic losses. In order to find effective and complex measures against this invasive plant, it is very important to identify places and areas where H. sosnowskyi grows, carry out a detailed analysis, and monitor its spread to avoid leaving this process to chance. In this paper, the remote sensing methodology was proposed to identify territories covered with H. sosnowskyi plants (land classification). Two categories of land cover classification were used: supervised (human-guided) and unsupervised (calculated by software). In the application of the supervised method, the average wavelength of the spectrum of H. sosnowskyi was calculated for the classification of the RGB image and according to this, the unsupervised classification by the program was accomplished. The combination of both classification methods, performed in steps, allowed obtaining better results than using one. The application of authors’ proposed methodology was demonstrated in a Lithuanian case study discussed in this paper.

scholarly journals Lead Detection in Polar Oceans—A Comparison of Different Classification Methods for Cryosat-2 SAR Data

2018 ◽  
Vol 10 (8) ◽  
pp. 1190 ◽  
Author(s):  
Denise Dettmering ◽  
Alan Wynne ◽  
Felix Müller ◽  
Marcello Passaro ◽  
Florian Seitz
Keyword(s):  
Unsupervised Classification ◽  
Detection Algorithm ◽  
Classification Method ◽  
Correct Identification ◽  
Polar Regions ◽  
Classification Methods ◽  
Lead Detection ◽  
Radar Echoes ◽  
Polar Oceans ◽  
Sar Data

In polar regions, sea-ice hinders the precise observation of Sea Surface Heights (SSH) by satellite altimetry. In order to derive reliable heights for the openings within the ice, two steps have to be fulfilled: (1) the correct identification of water (e.g., in leads or polynias), a process known as lead classification; and (2) dedicated retracking algorithms to extract the ranges from the radar echoes. This study focuses on the first point and aims at identifying the best available lead classification method for Cryosat-2 SAR data. Four different altimeter lead classification methods are compared and assessed with respect to very high resolution airborne imagery. These methods are the maximum power classifier; multi-parameter classification method primarily based on pulse peakiness; multi-observation analysis of stack peakiness; and an unsupervised classification method. The unsupervised classification method with 25 clusters consistently performs best with an overall accuracy of 97%. Furthermore, this method does not require any knowledge of specific ice characteristics within the study area and is therefore the recommended lead detection algorithm for Cryosat-2 SAR in polar oceans.

Classification methods for functional data

2018 ◽  
Author(s):  
Amparo Baillo ◽  
Antonio Cuevas ◽  
Ricardo Fraiman
Keyword(s):  
Functional Data ◽  
Supervised Classification ◽  
Nearest Neighbor ◽  
Real Data ◽  
Reproducing Kernels ◽  
Unsupervised Classification ◽  
Classification Problem ◽  
Classification Methods ◽  
K Nearest Neighbor ◽  
Empirical Risk

This article reviews the literature concerning supervised and unsupervised classification of functional data. It first explains the meaning of unsupervised classification vs. supervised classification before discussing the supervised classification problem in the infinite-dimensional case, showing that its formal statement generally coincides with that of discriminant analysis in the classical multivariate case. It then considers the optimal classifier and plug-in rules, empirical risk and empirical minimization rules, linear discrimination rules, the k nearest neighbor (k-NN) method, and kernel rules. It also describes classification based on partial least squares, classification based on reproducing kernels, and depth-based classification. Finally, it examines unsupervised classification methods, focusing on K-means for functional data, K-means for data in a Hilbert space, and impartial trimmed K-means for functional data. Some practical issues, in particular real-data examples and simulations, are reviewed and some selected proofs are given.

scholarly journals Quantitative and Qualitative Comparison Study between Classification Methods on normal and abnormal (with a coronavirus ) lung images

2021 ◽  
Vol 26 (3) ◽  
pp. 10-16
Author(s):  
Halaa Kadhim hasan ◽  
Ayad A.AL-Ani ◽  
Noor Z. AlKhazraji
Keyword(s):  
Unsupervised Classification ◽  
Classification Method ◽  
Classification Methods ◽  
Comparison Study ◽  
Qualitative Properties ◽  
Qualitative Comparison ◽  
Different Populations ◽  
Quantitative Properties ◽  
Analysis Of Performance ◽  
Better Than

Classification is concerned with establishing criteria that can be used to identify or distinguish different populations of objects that appear in images. In this paper Supervised and unsupervised classification method applied on normal, abnormal (with a coronavirus) ct- lung images (which it took from Al shaikh zaeid Hospital)  to study the quantitative and qualitative properties of these two categories. The analysis of performance with default quantitative parameters revealed that (kurtosis, skewness, entropy, Stander deviation (STD), mean). We found that: Qualitative (as seen) of   abnormal lung images after applying  Supervisors classification are better than the qualitative of abnormal lung images after applying  unsupervisors classification to detect the virus with white color in the lower lobes of the lung.. from The quantitative Properties such as (kurtosis, skewness) of original lung images are similar in rising to resulted value after applying  Supervisors classification on it, so Supervisors method is better than unSupervisors method to distinguishing between normal and abnormal lung images.

scholarly journals Discovery of a new song mode in Drosophila reveals hidden structure in the sensory and neural drivers of behavior

2017 ◽  
Author(s):  
Jan Clemens ◽  
Philip Coen ◽  
Frederic A. Roemschied ◽  
Talmo Pereira ◽  
David Mazumder ◽  
...  
Keyword(s):  
Sensory Feedback ◽  
Brain Activity ◽  
Unsupervised Classification ◽  
Time Varying ◽  
Classification Methods ◽  
Wing Vibration ◽  
Genetic Mechanisms ◽  
Song Production ◽  
Sensory Neural ◽  
Female Responses

SummaryDeciphering how brains generate behavior depends critically on an accurate description of behavior. If distinct behaviors are lumped together, separate modes of brain activity can be wrongly attributed to the same behavior. Alternatively, if a single behavior is split into two, the same neural activity can appear to produce different behaviors [1]. Here, we address this issue in the context of acoustic communication in Drosophila. During courtship, males utilize wing vibration to generate time-varying songs, and females evaluate songs to inform mating decisions [2-4]. Drosophila melanogaster song was thought for 50 years to consist of only two modes, sine and pulse, but using new unsupervised classification methods on large datasets of song recordings, we now establish the existence of at least three song modes: two distinct, evolutionary conserved pulse types, along with a single sine mode. We show how this seemingly subtle distinction profoundly affects our interpretation of the mechanisms underlying song production, perception and evolution. Specifically, we show that sensory feedback from the female influences the probability of producing each song mode and that male song mode choice affects female responses and contributes to modulating his song amplitude with distance [5]. At the neural level, we demonstrate how the activity of three separate neuron types within the fly’s song pathway differentially affect the probability of producing each song mode. Our results highlight the importance of carefully segmenting behavior to accurately map the underlying sensory, neural, and genetic mechanisms.

On Objective-Based Rough Hard and Fuzzyc-Means Clustering

2015 ◽  
Vol 19 (1) ◽  
pp. 29-35 ◽  
Author(s):  
Naohiko Kinoshita ◽  
◽  
Yasunori Endo ◽  
Keyword(s):  
Objective Function ◽  
Rough Set ◽  
Clustering Algorithms ◽  
Unsupervised Classification ◽  
Classification Methods ◽  
Clustering Methods ◽  
Clustering Method ◽  
Numerical Examples ◽  
Initial Values

Clustering is one of the most popular unsupervised classification methods. In this paper, we focus on rough clustering methods based on rough-set representation. Rough k-Means (RKM) is one of the rough clustering method proposed by Lingras et al. Outputs of many clustering algorithms, including RKM depend strongly on initial values, so we must evaluate the validity of outputs. In the case of objectivebased clustering algorithms, the objective function is handled as the measure. It is difficult, however to evaluate the output in RKM, which is not objective-based. To solve this problem, we propose new objective-based rough clustering algorithms and verify theirs usefulness through numerical examples.

scholarly journals Comparison of the classification methods for the images modeled by Gaussian random fields

2011 ◽  
Vol 52 ◽  
Author(s):  
Lijana Stabingienė ◽  
Giedrius Stabingis ◽  
Kęstutis Dučinskas
Keyword(s):  
Image Classification ◽  
Random Fields ◽  
Unsupervised Classification ◽  
Remotely Sensed ◽  
Gaussian Random Fields ◽  
Classification Methods ◽  
Remotely Sensed Image ◽  
Occurrence Matrix ◽  
Supervised Classification Methods ◽  
Correlation Range

In image classification often occur such situations, when images in some level are corrupted by additive noise. Such noise in image classification can be modeled by Gaussian random fields (GRF). In image classification supervised and unsupervised methods are used. In this paper we compare our proposed supervised classification methods based on plugin Bayes discriminant functions (PBDF) (see [6] and [11]) with unsupervised classification method based on grey level co-occurrence matrix (GLCM) (see e.g. [8] and [1]). The remotely sensed image is used for classification (USGS Earth Explorer). Also GRF with different spatial correlation range are generated and added to the original remotely sensed image. Such situation can naturally occur during forest fire, when smoke covers some territory. These images are used for classification accuracy examination.  

Sign in / Sign up

Export Citation Format

Share Document