Pattern Recognition

Pattern recognition stems from the need for automated machine recognition of objects, signals or images, or the need for automated decision-making based on a given set of parameters. Despite over half a century of productive research, pattern recognition continues to be an active area of research because of many unsolved fundamental theoretical problems as well as a rapidly increasing number of applications that can benefit from pattern recognition. A fundamental challenge in automated recognition and decision-making is the fact that pattern recognition problems that appear to be simple for even a 5-year old may in fact be quite difficult when transferred to machine domain. Consider the problem of identifying the gender of a person by looking at a pictorial representation. Let us use the pictures in Fig. 1 for a simple demonstration, which include actual photographs as well as cartoon renderings. It is relatively straightforward for humans to effortlessly identify the genders of these people, but now consider the problem of having a machine making the same decision. What distinguishing features are there between these two classes—males and females—that the machine should look at to make an intelligent decision? It is not difficult to realize that many of the features that initially come to mind, such as hair length, height-to-weight ratio, body curvature, facial expressions, or facial bone structure— even when used in combination—may fail to provide correct male/female classification of these images. Although we can naturally and easily identify each person as male or female, it is not so easy to determine how we come to this conclusion, or more specifically, what features we use to solve this classification problem. Of course, real-world pattern recognition problems are considerably more difficult then even the one illustrated above, and such problems span a very wide spectrum of applications, including speech recognition (e.g., automated voice-activated customer service), speaker identification, handwritten character recognition (such as the one used by the postal system to automatically read the addresses on envelopes), topographical remote sensing, identification of a system malfunction based on sensor data or loan/credit card application decision based on an individual’s credit report data, among many others. More recently, a growing number of biomedical engineeringrelated applications have been added to this list, including DNA sequence identification, automated digital mammography analysis for early detection of breast cancer, automated electrocardiogram (ECG) or electroencephalogram (EEG) analysis for cardiovascular or neurological disorder diagnosis, and biometrics (personal identification based on biological data such as iris scan, fingerprint, etc.). The list of applications can be infinitely extended, but all of these applications share a common denominator: automated classification or decision making based on observed parameters, such as a signal, image, or in general a pattern, obtained by combining several observations or measurements. This article provides an introductory background to pattern recognition and is organized as follows: The ter-