Natural Language Processing and Machine Vision

Natural Language Processing:

The natural Language Processing or in short NLP refers to the AI method of communicating with an intelligent system while using a natural language such as English. The processing of natural Language is required when we want an intelligent system like the robot to perform as per our instructions or when we want to hear a decision from a dialogue based clinical expert system and much more.

The field of NLP involves making computers perform useful tasks with the natural languages that we humans use. The input and output of an NLP system can be:

• Speech
• Written Text.

Components of NLP:

There are two components of NLP as given below:

1. Natural Language Understanding (NLU): The natural language understanding involves the tasks such as mapping the given input in natural language into useful representations and the analysis of different aspects of the language.
   
   
2. Natural Language Generation (NLG): It is the process of producing the meaningful phrases and the sentences in the form of natural language from some of the internal representation. It involves:
   Text planning: It includes retrieving the relevant content from the knowledge base.
   Sentence planning: It includes choosing the required words, forming the meaningful phrases, setting the tone of the sentence.
   Text Realization: It is the mapping of the sentence plan into the sentence structure.

The NLU is harder than the NLG.

Steps in NLP

There are generally five steps that are involved in natural language processing and they are given below:

• Lexical Analysis: It involves identifying and analyzing the structure of words. The Lexicon of a language means the collection of words and phrases in a language. The lexical analysis is dividing the whole chunk of text into paragraphs, sentences and words.
• Syntactic Analysis/Parsing: It involves the analysis of words in the sentence for grammar and arranging those words in such a manner that it shows the relationship between the words. The sentence such as “The school goes to boy” is rejected by English syntactic analyzer.
• Semantic Analysis: This analysis draws either the exact meaning or the dictionary meaning from the text. The text is checked for a quality of significance. It is done by mapping the syntactic structures along with the objects in the task domain. The semantic analyzer disregards the sentence such as “hot ice-cream”.
• Discourse Integration: The meaning of any sentence depends on upon the meaning of the sentence just before it. In addition, it also brings about the meaning of immediately succeeding sentence.
• Pragmatic Analysis: During this analysis what was said is re-interpreted on what it actually meant. It involves driving those aspects of language which require the real world knowledge.

Machine Vision:

Machine vision is the ability of a computer to "see". The concept of seeing involves a number of different tasks such as recognition of objects, tracking of objects, interpretation of scenes and objects and detection of new activities. Machine vision is also used for automatic recovery of 3-dimensional models of the environment.

Machine vision is considered to be sub-field of artificial intelligence. AI studies the computational aspects of intelligence. There are two important specifications in any vision system and they are the sensitivity and the resolution. The sensitivity is simply understood as the ability of a machine to see in dim light or to detect the weak impulses at invisible wavelengths. The resolution is referred as the extent to which a machine can differentiate between the objects. In general, the better the resolution, the more confined the field of vision. Sensitivity and resolution are interdependent. All other factors held the constant such as increasing the sensitivity reduces the resolution and improving the resolution reduces the sensitivity.

The applications of the machine vision system are presented below:

• Robotics:Machine vision can make a robot manipulator much more versatile by allowing it to deal with variations in parts, position, and orientation.
• Medicine: It assists a physician to reach a diagnosis. It does this by constructing 2D or 3D anatomy models of the human body. It analyzes the image to extract useful features.
• Remote Sensing: It senses the remote by taking the images from high altitudes such as from air-crafts or satellites. It analyses the image by generating a description.
• Cartography
• Meteorology
• Quality inspection
• Reconnaissance.

There is no universal machine vision system as there is only one system for each application.

References:

1. Elaine Rich, Kevin Knight 1991, "Artificial Intelligence".
2. Nilsson, Nils J. Principles of Artificial Intelligence, Narosa Publishing House New Delhi, 1998.
3. Norvig, Peter & Russel, Stuart Artificial Intelligence: A modern Approach, Prentice Hall, NJ, 1995
4. Patterson, Dan W. Introduction to Artificial Intelligence and Expert Systems, Prentice Hall of India Private Limited New Delhi, 1998.