Presentation on theme: "UKCI05 5-7 September 1 Applicability of Fuzzy Clustering for the Identification of Upwelling Areas on Sea Surface Temperature Images Susana Nascimento,"— Presentation transcript:
UKCI05 5-7 September 1 Applicability of Fuzzy Clustering for the Identification of Upwelling Areas on Sea Surface Temperature Images Susana Nascimento, Fátima M. Sousa, Hugo Casimiro Dmitri Boutov 2 Instituto de Oceanografia Faculdade de Ciências Universidade de Lisboa, PORTUGAL 1 Centro de Inteligência Artificial Dep. Informática Faculdade de Ciências e Tecnologia Universidade Nova de Lisboa PORTUGAL
UKCI05 5-7 September 2 Overview Introduction to the problem of Upwelling Recognition Sea Surface Temperature (SST) Image Segmentation by Fuzzy Partitional Clustering Methodology Experimental Study Ongoing Work
UKCI05 5-7 September 3 Upwelling Event What is Upwelling? It is a mass of deep, cold, and nutrient-rich seawater that rises close to the coast. Upwelling occurs when winds parallel to the coast induce a net mass transport of surface seawater in a 90º direction, away from the coast, due to the Coriolis force. Deep waters rise in order to compensate the mass deficiency that develops along the coastal area. Why is Upwelling so important? Brings nutrient-rich deep waters close to the ocean surface, creating regions of high biological productivity. Strong impact on fisheries, and global oceanic climate models http://oceanexplorer.noaa.gov/explorations/02quest/background/upwelling/upwelling.html
UKCI05 5-7 September 4 Upwelling Event in the Coastal Waters of Portugal SST image of an upwelling event obtained on 04AUG1998 (n14_98216_0422_sst); (b) upwelling boundary manually contoured; (c) upwelling areas automatically retrieved. Ground truth image
UKCI05 5-7 September 5 Why an Automatic System for Upwelling Recognition? Satellite Station of Instituto de Oceanografia (IO) of FC-UL Reception AVHRR thermal infrared Images since 1991 100 images per Upwelling Epoc (June-September) An expert chooses, by visual inspection, the best image of a day reception and treatment of 3-4 images a day. Until now, the areas covered by upwelling waters including cold filaments, have been contoured by hand. The method is very subjective and depending on the skill and practice of the expert.
UKCI05 5-7 September 6 Data AVHRR thermal infrared images are received and processed by IO Station with SeaSpace software package TeraScan producing SST images. Sea Surface Temperature (SST) images 720 400 matrix with each entry a temperature value in degrees Celsius with 1Km 2 spatial resolution. X Y
UKCI05 5-7 September 7 Distinct Groups of Images (G1) well-defined upwelling events (G2) images where upwelling is evident but there are areas with no temperature information (covered with clouds or noise); SST images divided into 5 groups according to different upwelling situations. (G5) Images lacking the upwelling event (G4) 3-day sequence of an upwelling event (G3) Upwelling event not well-defined;
UKCI05 5-7 September 8 Nature of the problem is Fuzzy Unsupervised segmentation does not require training data. Expert´s can take advantage of visualization skills and interpretability of fuzzy membership values. Why SST Image Segmentation by Fuzzy Clustering? Upwelling frontier
UKCI05 5-7 September 10 Fuzzy Clustering k-means vs Fuzzy c-means FCM AO Algorithms Fuzzy c-Means (FCM) Validity Guided (re)Clustering Adaptive variants... Parameters 1. sharpness exponent m, 2. number of clusters c FCM Features Data representation: objects are vectors of measured values. Clusters shape: different geometric prototypes; norms or scalar products. Clusters size: use of adaptive distance or adaptive algorithms. Clusters validity: optimal number of classes through validity functionals, clusters merging/splitting or by using a hierarchical approach. Final fuzzy partition: can be defuzzied; fuzzy partition should not be discarded Method: fuzzy objective function minimization; two step iterative procedure that continually decreases the value of the objective function FCM Features Data representation: objects are vectors of measured values. Clusters shape: different geometric prototypes; norms or scalar products. Clusters size: use of adaptive distance or adaptive algorithms. Clusters validity: optimal number of classes through validity functionals, clusters merging/splitting or by using a hierarchical approach. Final fuzzy partition: can be defuzzied; fuzzy partition should not be discarded Method: fuzzy objective function minimization; two step iterative procedure that continually decreases the value of the objective function
UKCI05 5-7 September 11 Spatial Visualization of Fuzzy c-Partition U=[u ik ] max membership value (u ik, i)
UKCI05 5-7 September 12 Ground truth image Oceanographer´s evaluation Assessment Accuracy Assessment cluster validation c XB SST image c MR Matching rate cEcE Fuzzy segmentation + visualization module Image matching c=2 c=3 c=4
UKCI05 5-7 September 13 Clustering Validation Small values of XB for compact and well- separated clusters. Xie-Beni (compactness and sepation) Index Other validation indexes Partition coeficient Partition entropy Davies-Bouldi... Other Validation approaches Adaptive algorithms totally unsupervised...
UKCI05 5-7 September 14 Consider two c-partitions P (1), P (2) of X 1.Maximal intersection 2.Matching rate of mapping P (1) P (2) 3.Matching rate of mapping P (2) P (1) 4.Matching rate, MR Image Matching 1. Defuzzify c-partition 2. Merge clusters 3. Measure matching rate Compare segmented and ground- truth images.
UKCI05 5-7 September 15 Experimental Study Main Goal To identify the upwelling event using fuzzy clustering analyse the enhancement of the upwelling areas To evaluate the number of clusters that better identifies the phenomena in a SST image. validation index To evaluate how closely the obtained segmentation reproduces the shape of the areas covered with upwelling waters. matching rate between fuzzy c-partition of SST and corresponding ground truth.
UKCI05 5-7 September 16 Experimental Study Reception of AVHRR thermal infrared Images Selection of SST Images and provide GT Images Image pre-processing Normalization Fuzzy Clustering Image Segmentation c=2, 3,..., 4 Ground truth assessment Clustering Validation Fuzzy partition Visualization Oceanographer´s Evaluation Used 16 SST images for all five groups represented Change in the mean temperature of the main clusters is not significant beyond four clusters (i.e. c > 4). for each c the FCM had been run from 10 distinct initialisations with sharpness parameter m= 2.0.
UKCI05 5-7 September 17 Summary of Results The FCM c-partitions for c=3, c=4 very closely represent the upwelling areas for all images of groups G 1, G 2, G 3, G 4 The upwelling areas correspond to the subset of clusters with the lowest mean temperatures The segmented results for the images with no upwelling, also lack the characteristic shape of the upwelling areas For 79% of segmented images, the FCM algorithm closely reproduces the shape of the areas covered with upwelling waters. The matching rate MR of selected partitions with GT images varied between 90% and 97%. The Xie-Beni index selects the correct number of clusters for 71% of images
UKCI05 5-7 September 18 Ongoing and Future Prospects Feature Selection o Temperature + spatial coordinates: no appearent improvments o Temperature + Distance to coast: an option Distinguish Upwelling from no-Upwelling Analysing the clusters of lowest mean temperature of two consecutive partitions P c, P c+1 : they split The behavior only occurs consistently for the days with Upwelling Spatio-temporal Analysis of Upwelling Events o Compare two consecutive partitions P c, P c+1 wrt o Mean temperature differences (i.e. cluster prototypes) o Change of membership assignment of points along the frontal boundaries o - cut analysis Hybridization of FCM + GA´s on cluster validation
UKCI05 5-7 September 19 Automatic Eddy Recognition and its Spatio-Temporal Tracking through Fuzzy Clustering Image Pre-processing to get edge enanhment o Image Filters + Normalization Feature extraction Segmentation using fuzzy clustering o e.g. Gath-Geva algorithm Developing Dynamical versions of Fuzzy Clustering and their adaptation to model Eddy Tracking Eddies are energetic swirling currents found all over the ocean o any temperature o distinct shapes
UKCI05 5-7 September 20 Remote Detection of Mediterranean Water Eddies in the Northeast Atlantic (RENA) RENA Project Funding Fundação para a Ciência e Tecnologia (FCT) European Space Agency (ESA)
UKCI05 5-7 September 22 Fuzzy c-Means Clustering Membership Values Weighted Fuzzy c-Means distance degree of fuzzification constraint Stepest descent constraint AO Algorithm Optimization of the performance index weight Given c= # of groups
UKCI05 5-7 September 24 Objective Automatic Identification of Eddy Patterns in Remote Sensed Satellite Images. Problem Illustration
UKCI05 5-7 September 25 Architecture Pre-Processing Fuzzy Clustering Histogram Feature Extraction Feature Selection ANN Classifier Training Evolutionary Algorithm Embedded Approach Structural (i.e. shape, orientation, size) oceanographic properties SegmentationClassification Spiral Description ? Windowing SOM Law´s method Oriented gradients Histogram Grid method Data Quantization Data Filtering
UKCI05 5-7 September 26 Task: Fuzzy Segmentation Unsupervised segmentation does not require training data Linguistic / visualization interpretability of fuzzy membership functions by the experts. Rule-based Segmentation Extraction of Fuzzy IF-THEN rules
UKCI05 5-7 September 27 Why Fuzzy Image Segmentation? Fuzzy membership functions provide natural means to model the ambiguity of patterns present in these images. n12_01104_0602 What is a segment ?
UKCI05 5-7 September 28 Histogram Spatial connectedness Grid method Data Quantization Region quantization Data points aggregation central value
UKCI05 5-7 September 29 Compressed Image by histogram
UKCI05 5-7 September 30 Fuzzy c-Means Clustering Membership Values Weighted Fuzzy c-Means distance degree of fuzzification constraint Stepest descent constraint AO Algorithm Optimization of the performance index weight Given c= # of groups
UKCI05 5-7 September 32 Original image Fuzzy Membership by thresholdingMax Fuzzy Membership Partition Defuzzified Partition
UKCI05 5-7 September 33 Evaluate Segmentation Quality Goal: Accurate quantitative evaluation of image Segmentations. Detection Accuracy: matching between reference optimal segmentation of ground-truth eddies and segmented ones. Select Validity Functional
UKCI05 5-7 September 34 Validity-Oriented Clustering Two main problems (P1) Objective function may not be a good estimator of true classification quality (as defined by the expert) (P2) Objective function often admits many suboptimal solutions. Strategy algorithm that evaluates generated partitions by a quality measure Modify bad partitions and improve their quality
UKCI05 5-7 September 35 Ongoing Work 1. Study of techniqes to evaluate segmentation quality. 2. Segmentation from other feature vectors. 3. Development of a totaly unsupervised FCM algorithm the number of clusters is determined by a validation functional. Validity measure based on cluster compactness and separation