Browsing by Author "Assabie, Yaregal(PhD)"

Now showing 1 - 14 of 14
  • No Thumbnail Available
    Item
    Amharic Document Image Retrieval Using Lingustic Features
    (Addis Ababa University, 2011-10-21) Yeshambel, Tilahun; Assabie, Yaregal(PhD)
    The advent of modern computers play important roles in processing and managing electronic information that are found in the form of texts, images, audios and videos, etc. With the rapid development of computer technology, digital documents have become popular options for storage, accessing and transmission. With the need of current fast evolving digital libraries, an increasing amount of historical documents, newspaper, books, etc. are being digitized into an electronic format for easy archival and dissemination purposes. Optical Character Recognition (OCR) and Document Image Retrieval (DIR), as part of information retrieval paradigm, are the two means of accessing document images that received attention among the IR community. Amharic is the official language of Ethiopia since 19th century and as a result so many religious and government documents are written in Amharic. Huge collections of Amharic machine printed documents are found in almost every institution of the country. It is observed that accessing those documents has become more and more difficult. To address this problem, very few number of research works have been attempted recently by using OCR and DIR methods. The aim of this research is to develop a system model that enables users to find relevant Amharic document images from a corpus of digitized documents in an easy, accurate, fast and efficient manner. So this work presents the architecture of Amharic DIR which allows users to search scanned Amharic documents without the need of OCR. The proposed model is designed after making detailed analysis of the specific nature of Amharic language. Amharic belongs to the Semitic languages and is morphologically rich language. Surface words formation involves prefixation, suffixation, infixation, circumfixation and reduplication. In this work a model for searching Amharic document images is proposed and word image features are systematically extracted for automatically indexing, retrieving and ranking of document images stored in a database. A new approach that applies one of the NLP tools which is Amharic word generator is incorporated in the proposed system model. By providing a given Amharic root word to this Amharic specific surface word synthesizer, a number of possible surface words are produced. Then, the descriptions of these surface word images are used for indexing and searching purposes. On the other hand the system passes through various phases such as noise removal, binirization, text line and word boundary identification, word segmentation and resizing to normalize different font types, sizes and styles, feature extraction and finally matching query word image against document word images. The proposed method was tested on different real world Amharic documents from different sources like magazines, textbooks and newspapers with various font styles, types and sizes. Precision-recall measures of evaluation had been conducted for sample queries on sample document images and promising results have been achieved.
  • No Thumbnail Available
    Item
    Amharic Document Image Retrieval Using Lingustic Features
    (Addis Ababa University, 10/21/2011) Yeshambel, Tilahun; Assabie, Yaregal(PhD)
    The advent of modern computers play important roles in processing and managing electronic information that are found in the form of texts, images, audios and videos, etc. With the rapid development of computer technology, digital documents have become popular options for storage, accessing and transmission. With the need of current fast evolving digital libraries, an increasing amount of historical documents, newspaper, books, etc. are being digitized into an electronic format for easy archival and dissemination purposes. Optical Character Recognition (OCR) and Document Image Retrieval (DIR), as part of information retrieval paradigm, are the two means of accessing document images that received attention among the IR community. Amharic is the official language of Ethiopia since 19th century and as a result so many religious and government documents are written in Amharic. Huge collections of Amharic machine printed documents are found in almost every institution of the country. It is observed that accessing those documents has become more and more difficult. To address this problem, very few number of research works have been attempted recently by using OCR and DIR methods. The aim of this research is to develop a system model that enables users to find relevant Amharic document images from a corpus of digitized documents in an easy, accurate, fast and efficient manner. So this work presents the architecture of Amharic DIR which allows users to search scanned Amharic documents without the need of OCR. The proposed model is designed after making detailed analysis of the specific nature of Amharic language. Amharic belongs to the Semitic languages and is morphologically rich language. Surface words formation involves prefixation, suffixation, infixation, circumfixation and reduplication. In this work a model for searching Amharic document images is proposed and word image features are systematically extracted for automatically indexing, retrieving and ranking of document images stored in a database. A new approach that applies one of the NLP tools which is Amharic word generator is incorporated in the proposed system model. By providing a given Amharic root word to this Amharic specific surface word synthesizer, a number of possible surface words are produced. Then, the descriptions of these surface word images are used for indexing and searching purposes. On the other hand the system passes through various phases such as noise removal, binirization, text line and word boundary identification, word segmentation and resizing to normalize different font types, sizes and styles, feature extraction and finally matching query word image against document word images. The proposed method was tested on different real world Amharic documents from different sources like magazines, textbooks and newspapers with various font styles, types and sizes. Precision-recall measures of evaluation had been conducted for sample queries on sample document images and promising results have been achieved.
  • No Thumbnail Available
    Item
    Automatic Flower Disease Identification Using Image Processing
    (Addis Ababa University, 2015-02) Tigistu, Getahun; Assabie, Yaregal(PhD)
    Currently, the cultivation of flowers is becoming popular. However, during the cultivation process there may be a number of challenges that affect it, one of which is flower disease. Most flower diseases are caused by insects, fungi, and bacteria. Identification of these diseases need experienced experts in this area. Thus, developing a system that automatically identifies flower diseases can help to support the experienced experts. In view of this, an image processing based system for automatic identification of flower disease is proposed. The proposed system consists of two main phases. In the first phase normal and diseased flower image are used to create a knowledge base. During the creation of the knowledge base, images are pre-processed and segmented to identify the region of interest. Then, seven different texture features of images are extracted using Gabor texture feature extraction. Finally, an artificial neural network is trained using seven input features extracted from the individual image and eight output vectors that represent eight different classes of disease to represent the knowledge base. In the second phase, the knowledge base is used to identify the disease of a flower. In order to create the knowledge base and to test the effectiveness of the developed system, we have used 40 flower images for each of the eight different classes of flower disease and we have a total of 320 flower images. From those images 85% of the Dataset is used for training and 15% of the data set is used for testing. The experimental result demonstrates that the proposed technique is effective technique for the identification of flower disease. The developed system can successfully identify the examined flower with an accuracy of 83.3%. Keywords: Gabor Feature Extraction, Artificial Neural Network, Texture Feature
  • No Thumbnail Available
    Item
    Automatic Flower Disease Identification Using Image Processing
    (Addis Ababa University, 2015-02) Tigistu, Getahun; Assabie, Yaregal(PhD)
    Currently, the cultivation of flowers is becoming popular. However, during the cultivation process there may be a number of challenges that affect it, one of which is flower disease. Most flower diseases are caused by insects, fungi, and bacteria. Identification of these diseases need experienced experts in this area. Thus, developing a system that automatically identifies flower diseases can help to support the experienced experts. In view of this, an image processing based system for automatic identification of flower disease is proposed. The proposed system consists of two main phases. In the first phase normal and diseased flower image are used to create a knowledge base. During the creation of the knowledge base, images are pre-processed and segmented to identify the region of interest. Then, seven different texture features of images are extracted using Gabor texture feature extraction. Finally, an artificial neural network is trained using seven input features extracted from the individual image and eight output vectors that represent eight different classes of disease to represent the knowledge base. In the second phase, the knowledge base is used to identify the disease of a flower. In order to create the knowledge base and to test the effectiveness of the developed system, we have used 40 flower images for each of the eight different classes of flower disease and we have a total of 320 flower images. From those images 85% of the Dataset is used for training and 15% of the data set is used for testing. The experimental result demonstrates that the proposed technique is effective technique for the identification of flower disease. The developed system can successfully identify the examined flower with an accuracy of 83.3%. Keywords: Gabor Feature Extraction, Artificial Neural Network, Texture Feature
  • No Thumbnail Available
    Item
    Automatic Recognition of Ethiopian Paper Currency
    (Addis Ababa University, 2014-10) Fentahun, Jegnaw; Assabie, Yaregal(PhD)
    Currency recognition is an image processing technology that is used to identify currency of various countries. Due to the use of currency in day to day life, the importance for automatic methods for currency recognition has been increasing. An efficient currency recognition system is vital for automation in many sectors such as vending machine, railway ticket counter, banking system, shopping mall, currency exchange service, etc. Due to this, automatic currency recognition has been the interest of many researchers and currency recognition was done for different countries’ currencies such as United States (US) dollar, Euro, Chinese Renminbi (RMN), Indian rupee and Mexican peso. However, to the best knowledge of the researcher, there is no any research done towards designing and implementing recognition of Ethiopian currency. The absence of such currency recognition is a big gap in Ethiopia. This thesis describes the design of automatic recognition of Ethiopian currency. In this research, a software solution which takes the image of an Ethiopian currency from a scanner and camera as an input is proposed. The researcher combined the approaches of currency characteristic comparison and local feature descriptors to design a four level classifier. The design has a categorization component, which is responsible to denominate the currency notes into their respective denomination and verification component which is responsible to validate whether the currency is genuine or not. Both components of the design are implemented using MATLAB. The design is tested using genuine Ethiopian currencies at different condition, counterfeit Ethiopian currencies and other countries’ currencies. The denomination accuracy for genuine Ethiopian currency, counterfeit currencies and other countries’ currencies is found to be 90.42%, 83.3% and 100% respectively. The verification accuracy is 96.13%. The overall processing time of the model is 1. 986 second. Therefore our model has a good performance with a denomination and verification accuracy more than 90%. KEY WORDS: Image Processing, Currency Recognition, Speeded Up Robust Feature (SURF) Feature, MATLAB, Hue Saturation Value (HSV), Counterfeit Detection, Classifier, Feature Extraction.
  • No Thumbnail Available
    Item
    Automatic Recognition of Ethiopian Paper Currency
    (Addis Ababa University, 2014-10) Fentahun, Jegnaw; Assabie, Yaregal(PhD)
    Currency recognition is an image processing technology that is used to identify currency of various countries. Due to the use of currency in day to day life, the importance for automatic methods for currency recognition has been increasing. An efficient currency recognition system is vital for automation in many sectors such as vending machine, railway ticket counter, banking system, shopping mall, currency exchange service, etc. Due to this, automatic currency recognition has been the interest of many researchers and currency recognition was done for different countries’ currencies such as United States (US) dollar, Euro, Chinese Renminbi (RMN), Indian rupee and Mexican peso. However, to the best knowledge of the researcher, there is no any research done towards designing and implementing recognition of Ethiopian currency. The absence of such currency recognition is a big gap in Ethiopia. This thesis describes the design of automatic recognition of Ethiopian currency. In this research, a software solution which takes the image of an Ethiopian currency from a scanner and camera as an input is proposed. The researcher combined the approaches of currency characteristic comparison and local feature descriptors to design a four level classifier. The design has a categorization component, which is responsible to denominate the currency notes into their respective denomination and verification component which is responsible to validate whether the currency is genuine or not. Both components of the design are implemented using MATLAB. The design is tested using genuine Ethiopian currencies at different condition, counterfeit Ethiopian currencies and other countries’ currencies. The denomination accuracy for genuine Ethiopian currency, counterfeit currencies and other countries’ currencies is found to be 90.42%, 83.3% and 100% respectively. The verification accuracy is 96.13%. The overall processing time of the model is 1. 986 second. Therefore our model has a good performance with a denomination and verification accuracy more than 90%. KEY WORDS: Image Processing, Currency Recognition, Speeded Up Robust Feature (SURF) Feature, MATLAB, Hue Saturation Value (HSV), Counterfeit Detection, Classifier, Feature Extraction.
  • No Thumbnail Available
    Item
    Bidirectional English – Afaan Oromo Machine Translation Using Hybrid Approach
    (Addis Ababa University, 2013-11) Daba, Jabesa; Assabie, Yaregal(PhD)
    Machine translation is one of the applications of natural language processing that studies the use of computer programs and software to translate one natural language into another in the form of text or speech. Since there is a need for translation of documents between English and Afaan Oromo languages there needs to be a mechanism to do so. Thus, this study resulted in the development of a bidirectional English-Afaan Oromo machine translation system using a hybrid approach. The research work is implemented using a hybrid of rule based and statistical approaches. Since English and Afaan Oromo have different sentence structures, we implement syntactic reordering approach which makes the structure of source sentences to be similar to the structure of target sentences. So, reordering rules are developed for simple, interrogative and complex English and Afaan Oromo sentences. In order to achieve the objective of this research work, a corpus is collected from different domain and prepared in a format suitable for use in the development process and classified as training set and test set. The reordering rules are applied on both the training and test sets in a preprocessing step. Since the system is bidirectional, two language models are developed; one for English and the other for Afaan Oromo. Translation models which assign a probability that a given source language text generates a target language text are built and a decoder which searches for the shortest path is used. Two major experiments are conducted by using two different approaches and their results are recorded. The first experiment is carried out by using a statistical approach. The result obtained from the experiment has a BLEU score of 32.39% for English to Afaan Oromo translation and 41.50% for Afaan Oromo to English translation. The second experiment is carried out by using a hybrid approach and the result obtained has a BLEU score of 37.41% for English to Afaan Oromo translation and 52.02% for Afaan Oromo to English translation. From the result, we can see that the hybrid approach is better than the statistical approach for the language pair and a better translation is acquired when Afaan Oromo is used as a source language and English is used as a target language. Key words: Machine Translation, Statistical Machine Translation, Hybrid Machine Translation, Reordering rule
  • No Thumbnail Available
    Item
    Bidirectional English – Afaan Oromo Machine Translation Using Hybrid Approach
    (Addis Ababa University, 2013-11) Daba, Jabesa; Assabie, Yaregal(PhD)
    Machine translation is one of the applications of natural language processing that studies the use of computer programs and software to translate one natural language into another in the form of text or speech. Since there is a need for translation of documents between English and Afaan Oromo languages there needs to be a mechanism to do so. Thus, this study resulted in the development of a bidirectional English-Afaan Oromo machine translation system using a hybrid approach. The research work is implemented using a hybrid of rule based and statistical approaches. Since English and Afaan Oromo have different sentence structures, we implement syntactic reordering approach which makes the structure of source sentences to be similar to the structure of target sentences. So, reordering rules are developed for simple, interrogative and complex English and Afaan Oromo sentences. In order to achieve the objective of this research work, a corpus is collected from different domain and prepared in a format suitable for use in the development process and classified as training set and test set. The reordering rules are applied on both the training and test sets in a preprocessing step. Since the system is bidirectional, two language models are developed; one for English and the other for Afaan Oromo. Translation models which assign a probability that a given source language text generates a target language text are built and a decoder which searches for the shortest path is used. Two major experiments are conducted by using two different approaches and their results are recorded. The first experiment is carried out by using a statistical approach. The result obtained from the experiment has a BLEU score of 32.39% for English to Afaan Oromo translation and 41.50% for Afaan Oromo to English translation. The second experiment is carried out by using a hybrid approach and the result obtained has a BLEU score of 37.41% for English to Afaan Oromo translation and 52.02% for Afaan Oromo to English translation. From the result, we can see that the hybrid approach is better than the statistical approach for the language pair and a better translation is acquired when Afaan Oromo is used as a source language and English is used as a target language. Key words: Machine Translation, Statistical Machine Translation, Hybrid Machine Translation, Reordering rule
  • No Thumbnail Available
    Item
    Design and Development of Part-of-speech Tagger for Kafi-noonoo Language
    (Addis Ababa University, 2013-11) Mekuria, Zelalem; Assabie, Yaregal(PhD)
    Part-Of-Speech tagger is a program that reads text in given language and assigns parts-of-speech such as noun, verb, adjective, etc. to each word and other token within the text. Several part-of-speech taggers are available on the web for different languages including Amharic, Oromifa and Tigrigna. However, these POS taggers cannot be applied directly for Kafi-noonoo language. Thus, this thesis presents a research work on Kafi-noonoo part-of-speech tagger. In order to develop the tagger, the study employed a hybrid approach i.e. HMM and rule-based tagger at sentence level. Developing part-of-speech tagger for a language has many advantages such as: it can be used as input for full parser; it can be used in text-to-speech system to correct the way of pronunciation, it can be used for surface linguistic analysis, it can be used as a pre-processing step for researchers who want to conduct higher level NLP application development and it also provide a way of learning the language by discovering the word category and grammar construction of the language. For training and testing purpose, 354 untagged Kafi-noonoo sentences are collected from two genres and annotated using an incremental corpus preparation approach. In addition to this, 34 part-of-speech tags are identified for tagging purpose. After assigning word class information on each word within the sentences, both HMM and rule-based taggers are trained on 90% of the tagged sentences to generate probabilities i.e. lexical and transitional probability for the statistical component of the hybrid tagger and set of transformation rules for the rule-based component of the hybrid tagger. Based on these probabilities and transformation rules, the hybrid tagger (combination of HMM and rule-based tagger) assigns the most suitable word class information for the given untagged Kafi-noonoo texts. The performance of the prototypes i.e. HMM, rule-based and hybrid taggers are tested using different experiments. As a result, HMM and rule-based tagger with unigram initial state tagger shows 77.19% and 61.88%accuracy respectively whereas, the hybrid tagger improve the accuracy to 80.47%. Key words: Part of speech tagger, HMM, Rule-based, Hybrid tagger and Transformation rules
  • No Thumbnail Available
    Item
    Design and Development of Part-of-speech Tagger for Kafi-noonoo Language
    (Addis Ababa University, 2013-11) Mekuria, Zelalem; Assabie, Yaregal(PhD)
    Part-Of-Speech tagger is a program that reads text in given language and assigns parts-of-speech such as noun, verb, adjective, etc. to each word and other token within the text. Several part-of-speech taggers are available on the web for different languages including Amharic, Oromifa and Tigrigna. However, these POS taggers cannot be applied directly for Kafi-noonoo language. Thus, this thesis presents a research work on Kafi-noonoo part-of-speech tagger. In order to develop the tagger, the study employed a hybrid approach i.e. HMM and rule-based tagger at sentence level. Developing part-of-speech tagger for a language has many advantages such as: it can be used as input for full parser; it can be used in text-to-speech system to correct the way of pronunciation, it can be used for surface linguistic analysis, it can be used as a pre-processing step for researchers who want to conduct higher level NLP application development and it also provide a way of learning the language by discovering the word category and grammar construction of the language. For training and testing purpose, 354 untagged Kafi-noonoo sentences are collected from two genres and annotated using an incremental corpus preparation approach. In addition to this, 34 part-of-speech tags are identified for tagging purpose. After assigning word class information on each word within the sentences, both HMM and rule-based taggers are trained on 90% of the tagged sentences to generate probabilities i.e. lexical and transitional probability for the statistical component of the hybrid tagger and set of transformation rules for the rule-based component of the hybrid tagger. Based on these probabilities and transformation rules, the hybrid tagger (combination of HMM and rule-based tagger) assigns the most suitable word class information for the given untagged Kafi-noonoo texts. The performance of the prototypes i.e. HMM, rule-based and hybrid taggers are tested using different experiments. As a result, HMM and rule-based tagger with unigram initial state tagger shows 77.19% and 61.88%accuracy respectively whereas, the hybrid tagger improve the accuracy to 80.47%. Key words: Part of speech tagger, HMM, Rule-based, Hybrid tagger and Transformation rules
  • No Thumbnail Available
    Item
    Recognition of Amharic Braille Using Direction Field
    (Addis Ababa University, 2011-06) Hassen, Miftah; Assabie, Yaregal(PhD)
    Invented by Louis Braille in 1829, braille has been a widely used means of written communication for the blind people around the world. Since its introduction in Ethiopia, there have been piles of Amharic braille documents produced by the blind people for different purposes. Since these documents are usually read and understood only by the same group of people, the blind, who have written them, the invaluable knowledge and information found there is highly restricted from reaching the sighted society. Automatic recognition of braille documents is therefore very instrumental in bridging the communication gab that exists between the blind and the sighted people. This thesis presents an approach for recognition of Amharic braille documents. We have used direction field tensor, which uses Gaussian filters and derivatives of Gaussians, for noise removal and isolation of braille dots from their background. We also designed a new skewness correction technique, which exploits the horizontal direction nature of braille dots. Attempts are made to determine braille dot sizes automatically so that the system works for different braille dot sizes. Braille character lines are constructed in order for subsequent operations to be performed only on the character lines. A half character detection method, which differentiates braille dots from noises is applied. Having recognized the half characters through analysis of their dot positions, the braille cells are formulated by way of examining horizontal distances between the half characters. We have finally used a lookup table for translating each of the formulated braille cells into their corresponding Amharic print characters. Braille documents, especially those that were used by previous works for comparison, are collected for experiment purpose. Our system achieved an average accuracy of 98.5%. 99.9% accuracy is achieved for good quality braille documents, while the accuracy for poor quality braille documents is 96.5%. The small errors observed in the experiment are attributed to some stains and defects present on braille documents.
  • No Thumbnail Available
    Item
    Recognition of Amharic Braille Using Direction Field
    (Addis Ababa University, 2011-06) Hassen, Miftah; Assabie, Yaregal(PhD)
    Invented by Louis Braille in 1829, braille has been a widely used means of written communication for the blind people around the world. Since its introduction in Ethiopia, there have been piles of Amharic braille documents produced by the blind people for different purposes. Since these documents are usually read and understood only by the same group of people, the blind, who have written them, the invaluable knowledge and information found there is highly restricted from reaching the sighted society. Automatic recognition of braille documents is therefore very instrumental in bridging the communication gab that exists between the blind and the sighted people. This thesis presents an approach for recognition of Amharic braille documents. We have used direction field tensor, which uses Gaussian filters and derivatives of Gaussians, for noise removal and isolation of braille dots from their background. We also designed a new skewness correction technique, which exploits the horizontal direction nature of braille dots. Attempts are made to determine braille dot sizes automatically so that the system works for different braille dot sizes. Braille character lines are constructed in order for subsequent operations to be performed only on the character lines. A half character detection method, which differentiates braille dots from noises is applied. Having recognized the half characters through analysis of their dot positions, the braille cells are formulated by way of examining horizontal distances between the half characters. We have finally used a lookup table for translating each of the formulated braille cells into their corresponding Amharic print characters. Braille documents, especially those that were used by previous works for comparison, are collected for experiment purpose. Our system achieved an average accuracy of 98.5%. 99.9% accuracy is achieved for good quality braille documents, while the accuracy for poor quality braille documents is 96.5%. The small errors observed in the experiment are attributed to some stains and defects present on braille documents.
  • No Thumbnail Available
    Item
    Recognition of Ethiopian Car Plate
    (Addis Ababa University, 2013-04) Nigussie, Seble; Assabie, Yaregal(PhD)
    As time goes, application areas of information and communication technologies are growing dramatically as well. In these days, every potential problem is automated or it is being automated in order to solve or make things better. In the area of transportation systems, a lot of applications have been developed with development of communication and information processing technologies, officially called Intelligent Transportation Systems (ITS). In this work, one of the fundamental elements of ITS called Car Plate Recognition (CPR) is developed for Ethiopian car plates. The proposed system has three major modules: Plate Detection, Character Segmentation and Character Recognition. For plate detection, a gabor filter based method is proposed. In this module, even though, the gabor filter is the core unit that roughly locates possible plate regions, the module also applies a series of other techniques, namely, binarization, morphological closing operation and connected component analysis consecutively on the filter response, to detect the legitimate plate region. For character segmentation process a connected component analysis method is used. But before, the actual segmentation process, the plate image passes through a number of preprocessing tasks that dramatically increase the accuracy of the segmentation outcome. Of these preprocessing tasks plate’s orientation correction and plate’s frame removal are the major ones. For plate’s orientation correction we used a combination of hough transform and shear transform. For plate’s frame elimination we used a series of binary operations. Besides these preprocessing tasks, the module performs further post segmentation operations that are done in the segmentation outcome of the CCA. The main objective of the post segmentation stage is to separate connected character objects (if any exists) with the help of plate structure information. Finally, a correlation based template matching method is used for character recognition. In addition to the correlation value, the recognition process is supported by color analysis techniques and location information of characters. The prototype of the proposed system is developed using MATLABTM and its performance is tested on 350 RGB car images that are taken under different angle, distance, motion and illumination conditions. The developed system results in an accuracy of 63.14% and also it is able to recognize a plate between 2-5 seconds depending on whether post processing operations are needed or not. Keywords: Ethiopian Car Plates, Ethiopian Car Plate Recognition, License Plate Recognition, Plate Detection, Plate Characters Segmentation, Character Recognition, Plate’s Frame Elimination, Ethiopian Car Plate Type Identification
  • No Thumbnail Available
    Item
    Recognition of Ethiopian Car Plate
    (Addis Ababa University, 2013-04) Nigussie, Seble; Assabie, Yaregal(PhD)
    As time goes, application areas of information and communication technologies are growing dramatically as well. In these days, every potential problem is automated or it is being automated in order to solve or make things better. In the area of transportation systems, a lot of applications have been developed with development of communication and information processing technologies, officially called Intelligent Transportation Systems (ITS). In this work, one of the fundamental elements of ITS called Car Plate Recognition (CPR) is developed for Ethiopian car plates. The proposed system has three major modules: Plate Detection, Character Segmentation and Character Recognition. For plate detection, a gabor filter based method is proposed. In this module, even though, the gabor filter is the core unit that roughly locates possible plate regions, the module also applies a series of other techniques, namely, binarization, morphological closing operation and connected component analysis consecutively on the filter response, to detect the legitimate plate region. For character segmentation process a connected component analysis method is used. But before, the actual segmentation process, the plate image passes through a number of preprocessing tasks that dramatically increase the accuracy of the segmentation outcome. Of these preprocessing tasks plate’s orientation correction and plate’s frame removal are the major ones. For plate’s orientation correction we used a combination of hough transform and shear transform. For plate’s frame elimination we used a series of binary operations. Besides these preprocessing tasks, the module performs further post segmentation operations that are done in the segmentation outcome of the CCA. The main objective of the post segmentation stage is to separate connected character objects (if any exists) with the help of plate structure information. Finally, a correlation based template matching method is used for character recognition. In addition to the correlation value, the recognition process is supported by color analysis techniques and location information of characters. The prototype of the proposed system is developed using MATLABTM and its performance is tested on 350 RGB car images that are taken under different angle, distance, motion and illumination conditions. The developed system results in an accuracy of 63.14% and also it is able to recognize a plate between 2-5 seconds depending on whether post processing operations are needed or not. Keywords: Ethiopian Car Plates, Ethiopian Car Plate Recognition, License Plate Recognition, Plate Detection, Plate Characters Segmentation, Character Recognition, Plate’s Frame Elimination, Ethiopian Car Plate Type Identification