Pattern Recognition and Classification Theory

plan Recognition and categorization TheoryAssignment 2 of Pattern recognition should contain the shed light onification theory. The topics should coverIntroduction to Pattern Recognition, including a) The concept of anatomy recognition and its applications. b) Basic steps of a typical descriptor recognition task. c) Popular techniques utilise in these steps. d) Various application areas of pattern recognition look for.bayian classification rule, prior, posterior, spill function, risk, and minimum error engraft up classification.Discriminant functions, Normal densities and application of Bayesian rule to normal densities with 3 distinguishable gaffes of variances and covariance matrices discussed in book.As the name indicates, the pattern recognition is the classification of a pattern to one of the pre-specified classes. The process of understanding or recognizing the patterns by pickings the vulgar information from a sensor, convert that raw data into roughly meta da ta by pre-process that raw data, producing segments of the data through approximately multifariousness of segmentation process and the pass those segment through some lark approximately extractor which lead the purification of the raw data to be understand by the Classifier. Based on the feature extracted will classify it to a certain class which is already defined by the conclusion landmark. The last boundary is obtained from a series of training data and the represent think to it.In short, the process of identifying object or pattern into some divide of classes based on some features which are been described by the decision rule. A simple example for it the identification of the seabass and salmon lean deprivation through a conveyer belt. Certain features uniform the height, width and luminosity can be used to develop a decision boundary and put any fish into its respected class (sea bass or salmon).It is the synopsis of in what way the machines observe the environ ment, come to know about the different patterns and make a rational decision about the class of the patterns.A typical pattern recognition system consists of the following components sensual EnvironmentData Acquisition/SensingPre-ProcessingFeature ExtractionFeaturesClassificationPost-ProcessingDecision MakingThe above mentioned components are presumption in the construe 1. direct 1 Components of a Pattern Recognition System.How to overcome the inadequateness of vector space?Numerous amount of training data.Anonymous distributions of classes.unnamed problem complexity.Generalization problems.Evaluation problems.Given below are fewer of the pattern recognition potential research areasAdaptive signal touch onMachine learningArtificial neural networksRobotics and visionCognitive sciences numeric statisticsNonlinear optimizationExploratory data analytic thinkingFuzzy and inheritable systemsDetection and estimation theoryFormal languagesStructural modelingbiological cyberneticsComp utational neurosciencePattern recognition has outnumbered amount of applications, some of which are as followsImage processingComputer visionSpeech recognitionMultimodal interfaces machine-driven target recognitionOptical character recognitionSeismic analysisMan and machine diagnosticsFingerprint identificationIndustrial revueFinancial forecastMedical diagnosisECG signal analysisGiven below are the fundamental steps involved in pattern recognitionSensing The pattern recognition systems require a sensor at the input in order to take raw data from the environment into the system.Segmentation It is done after the pre-processing step. In some systems this is the pre-processing step used for converting the raw data into some signifiered data for the feature extraction.Feature Extraction Some specific parameters of the pattern are measured in this step like length in the fish example.Classification The patterns are then classified through some sort of classifiers like Bayesian Classifie r. Classification is done for putting the pattern into a specific class or category e.g. sea bass or salmon.Post Processing This step is done for further improvement of the performance.Figure 2 Steps involved in Pattern Recognition.Classification techniques Bayes classifier, HMM, Kth Nearest Neighbor (KNN), Artificial Neural Network (ANN), Support vector Machines (SVM), Training (parameter finding) testing (decoding) etc.Data representation techniquesThe compacting technique is used for improving the characteristics features of data using various transformation methods like the Fourier metamorphose method, WT etc.Dimensionality reductionReduce the data dimensions by removing the mutually tally features which results in the reduction of the common information to produce a set of nearly real informative parameters.e.g. Principle Component outline, Linear Discriminant Analysis etc.TransformationsVarious transformation techniques are likewise used like Fourier Transforms, Fast Fou rier Transform etc.The following are the potential research areas in the field of pattern recognitionAdaptive signal processingMachine learningArtificial neural networksRobotics and visionCognitive sciences numeral statisticsNonlinear optimizationExploratory data analysisFuzzy and heritable systemsDetection and estimation theoryFormal languagesStructural modelingbiological cyberneticsComputational neuroscienceThe probability of a state of nature that show how apt(predicate) is that, that particular state of nature would occur. For example, in the fish example it is prone that the prior of the salmon is 0.85. This mean that salmon is 85% more seeming to appear than the sea bass. If number of classes are c, thenIt is the probability of a specific state of nature given that observables have occurred. Mathematically,Notice that,It shows the cost related to each wrong action or decision we take. Mathematically,The zero-one is the almost comm further used dismission function. It ass igns zero on no loss in case of correct decision while in case of incorrect decision, it takes a uniform unit loss. Mathematically,The expected loss is also called as conditional risk. It is defined as the summation of the product of loss occurred from each decision to its posterior probability. MathematicallyOverall risk is given byFrom above equation we come to know that by selecting only those action (.) that belittle the for all values of x will minimize the overall risk which is directly associated with the error thus minimize the error rate.