logging in or signing up Handwritten Character Recognition Using Artificial Intelligence anuj87420 Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT lite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 103 Category: Education License: All Rights Reserved Like it (0) Dislike it (0) Added: February 09, 2012 This Presentation is Public Favorites: 0 Presentation Description for seminar purpose Comments Posting comment... By: shridhar22 (3 month(s) ago) whats your email id dear??? Saving..... Post Reply Close By: anuj87420 (3 month(s) ago) first f all thanx for trusting me dude..........surely i wil never break ur trust!!!!!!!! my id is anuj_raj2009@yahoo.co.in thanx again for helping me in securing gud marks wid regards anuj raj Saving..... Edit Comment Close By: shridhar22 (3 month(s) ago) just answer me the truth... did u make this presentation yourself??? Answer the truth... and i'll send u my ppt on your email. Saving..... Post Reply Close By: anuj87420 (3 month(s) ago) hey yaar i hav seminar on skinput on monday........so its upto u that u wil send or not.......but surely i wil not forward 2 anyone if u wil not want.......so help me becoz my 50 marks depends only on seminar......i take 1 or 2 ppts from others but urs is best........so help me if u want & plse reply me either in positive or negative.... thanx fr ur concern anuj raj By: anuj87420 (3 month(s) ago) hey yaar i didnt make it myself but my frnd nitin in iit roorkee made it in his 2nd sem.....so, many frnds ask him for the ppt but he is busy wid his work so he told me 2 upload it publicly.......i also uploaded at scribd.com Saving..... Edit Comment Close Premium member Presentation Transcript Handwritten Character Recognition Using Artificial Intelligence: Handwritten Character Recognition Using Artificial Intelligence Submitted by Nitin Kumar Enroll no. 10528017 Guided by Dr. Saktidev Mukherjee Professor, Electrical Dept.Handwritten Character Recognition Using Artificial Intelligence: Handwritten Character Recognition Using Artificial Intelligence I will design three different Systems for handwritten character recognition system. Based on different types of features. Statistical features Structural features Global transformations and moments featuresOCR Systems: OCR Systems OCR systems consist of three major stages : Pre-processing Feature Extraction ClassificationPre-processing: Pre-processing The raw data is subjected to a number of preliminary processing steps to make it usable in the descriptive stages of character analysis. Pre-processing aims to produce data that are easy for the OCR systems to operate accurately. The main objectives of pre-processing are : Noise reduction Binarization Width normalization* Skew correction* Slant removal*Noise Reduction - Normalization: Noise Reduction - Normalization Normalization provides a tremendous reduction in data size, thinning extracts the shape information of the characters. Noise reduction improves the quality of the document. Two main approaches: Filtering (Median filter) Morphological Operations (erosion, dilation, etc)Binarization: Binarization Document image binarization (thresholding) refers to the conversion of a gray-scale image into a binary image. Two categories of thresholding: Global, picks one threshold value for the entire document image which is often based on an estimation of the background level from the intensity histogram of the image. Adaptive (local), uses different values for each pixel according to the local area informationSkew Correction: Skew Correction Skew Correction methods are used to align the paper document with the coordinate system of the scanner. Main approaches for skew detection include correlation, projection profiles.Slant Removal : Slant Removal The slant of handwritten texts varies from user to user. Slant removal methods are used to normalize the all characters to a standard form.Slant Removal : Slant Removal Entropy The dominant slope of the character is found from the slope corrected characters which gives the minimum entropy of a vertical projection histogram. The vertical histogram projection is calculated for a range of angles ± R. In our case R=60, seems to cover all writing styles. The slope of the character, ,is found from: The character is then corrected by using:Feature Extraction: Feature Extraction In feature extraction stage each character is represented as a feature vector, which becomes its identity. The major goal of feature extraction is to extract a set of features, which maximizes the recognition rate with the least amount of elements. Due to the nature of handwriting with its high degree of variability and imprecision obtaining these features, is a difficult task. Feature extraction methods are based on 3 types of features: Statistical features Structural features Global transformations and moments features1.Statistical Features: 1.Statistical Features Representation of a character image by statistical distribution of points takes care of style variations to some extent. The major statistical features used for character representation are: Zoning Projections and profilesZoning: Zoning The character image is divided into NxM zones. From each zone features are extracted to form the feature vector. The goal of zoning is to obtain the local characteristics instead of global characteristicsZoning – Density Features: Zoning – Density Features The number of foreground pixels, or the normalized number of foreground pixels, in each cell is considered a feature. Darker squares indicate higher density of zone pixels.Zoning – Direction Features : Zoning – Direction Features Based on the contour of the character image For each zone the contour is followed and a directional histogram is obtained by analyzing the adjacent pixels in a 3x3 neighborhoodZoning – Direction Features : Zoning – Direction Features Based on the skeleton of the character image Distinguish individual line segments Labeling line segment information Line segments are coded with a direction number 2 = vertical line segment 3 = right diagonal line segment 4 = horizontal line segment 5 = left diagonal line segment Formation of feature vector through zoning number of horizontal lines number of right diagonal lines number of vertical lines number of left diagonal lines number of intersection pointsProjection Histograms: Projection Histograms The basic idea behind using projections is that character images, which are 2-D signals, can be represented as 1-D signal. These features, although independent to noise and deformation, depend on rotation. Projection histograms count the number of pixels in each column and row of a character image. Projection histograms can separate characters such as “m” and “n” .Profiles: Profiles The profile counts the number of pixels (distance) between the bounding box of the character image and the edge of the character. The profiles describe well the external shapes of characters and allow to distinguish between a great number of letters, such as “p” and “q”.2.Structural Features: 2.Structural Features Characters can be represented by structural features with high tolerance to distortions and style variations. This type of representation may also encode some knowledge about the structure of the object or may provide some knowledge as to what sort of components make up that object. Structural features are based on topological and geometrical properties of the character, such as aspect ratio , cross points, loops, branch points, horizontal curves at top or bottom, etc.Structural Features: Structural FeaturesStructural Features: Structural Features A structural feature extraction method for recognizing handwritten characters Two types of features: Horizontal projection histograms Vertical projection histograms3.Global Transformations - Moments: 3.Global Transformations - Moments The Fourier Transform (FT) of the contour of the image is calculated. Since the first n coefficients of the FT can be used in order to reconstruct the contour, then these n coefficients are considered to be a n -dimesional feature vector that represents the character. Central moments that make the process of recognizing an object scale, translation, and rotation invariant . T he original image can be completely reconstructed from the moment coefficients.Classification: Classification ANN is used for classification purpose.Results so far…: Results so far…References: References [1] S. Suresh, S.N. Omkar, and V. Mani, " Parallel Implementation of Back-Propagation Algorithm in Networks of Workstations " IEEE Transactions on Parallel and Distributed Systems, vol. 16, no. 1,pp.24-34, January 2005 [2] Wani, M.A.; Rashid, S., “ Parallel Algorithm for Control Pattern Recognition ” , Fourth International Conference on Machine Learning and Applications, 15-17 Dec. 2005 IEEE. [3] Andrew T. Wilson, " Off-line Handwriting Recognition Using Artificial Neural Networks ", University of Minnesota, Morris, wilsonat@mrs.umn.edu [4] Robert Sabourin, Luiz E. Soares de Oliveira, Edouard Lethelier, Fl´avio Bortolozzi, " A New Segmentation Approach for Handwritten Digits “ [5] Michael Blumenstein and Brijesh Verma, " A Neural Network for Real-World Postal Address Recognition ", School of Information Technology, Faculty of Engineering and Applied Science, Griffith University, Australia You do not have the permission to view this presentation. In order to view it, please contact the author of the presentation.
Handwritten Character Recognition Using Artificial Intelligence anuj87420 Download Post to : URL : Related Presentations : Share Add to Flag Embed Email Send to Blogs and Networks Add to Channel Uploaded from authorPOINT lite Insert YouTube videos in PowerPont slides with aS Desktop Copy embed code: (To copy code, click on the text box) Embed: URL: Thumbnail: WordPress Embed Customize Embed The presentation is successfully added In Your Favorites. Views: 103 Category: Education License: All Rights Reserved Like it (0) Dislike it (0) Added: February 09, 2012 This Presentation is Public Favorites: 0 Presentation Description for seminar purpose Comments Posting comment... By: shridhar22 (3 month(s) ago) whats your email id dear??? Saving..... Post Reply Close By: anuj87420 (3 month(s) ago) first f all thanx for trusting me dude..........surely i wil never break ur trust!!!!!!!! my id is anuj_raj2009@yahoo.co.in thanx again for helping me in securing gud marks wid regards anuj raj Saving..... Edit Comment Close By: shridhar22 (3 month(s) ago) just answer me the truth... did u make this presentation yourself??? Answer the truth... and i'll send u my ppt on your email. Saving..... Post Reply Close By: anuj87420 (3 month(s) ago) hey yaar i hav seminar on skinput on monday........so its upto u that u wil send or not.......but surely i wil not forward 2 anyone if u wil not want.......so help me becoz my 50 marks depends only on seminar......i take 1 or 2 ppts from others but urs is best........so help me if u want & plse reply me either in positive or negative.... thanx fr ur concern anuj raj By: anuj87420 (3 month(s) ago) hey yaar i didnt make it myself but my frnd nitin in iit roorkee made it in his 2nd sem.....so, many frnds ask him for the ppt but he is busy wid his work so he told me 2 upload it publicly.......i also uploaded at scribd.com Saving..... Edit Comment Close Premium member Presentation Transcript Handwritten Character Recognition Using Artificial Intelligence: Handwritten Character Recognition Using Artificial Intelligence Submitted by Nitin Kumar Enroll no. 10528017 Guided by Dr. Saktidev Mukherjee Professor, Electrical Dept.Handwritten Character Recognition Using Artificial Intelligence: Handwritten Character Recognition Using Artificial Intelligence I will design three different Systems for handwritten character recognition system. Based on different types of features. Statistical features Structural features Global transformations and moments featuresOCR Systems: OCR Systems OCR systems consist of three major stages : Pre-processing Feature Extraction ClassificationPre-processing: Pre-processing The raw data is subjected to a number of preliminary processing steps to make it usable in the descriptive stages of character analysis. Pre-processing aims to produce data that are easy for the OCR systems to operate accurately. The main objectives of pre-processing are : Noise reduction Binarization Width normalization* Skew correction* Slant removal*Noise Reduction - Normalization: Noise Reduction - Normalization Normalization provides a tremendous reduction in data size, thinning extracts the shape information of the characters. Noise reduction improves the quality of the document. Two main approaches: Filtering (Median filter) Morphological Operations (erosion, dilation, etc)Binarization: Binarization Document image binarization (thresholding) refers to the conversion of a gray-scale image into a binary image. Two categories of thresholding: Global, picks one threshold value for the entire document image which is often based on an estimation of the background level from the intensity histogram of the image. Adaptive (local), uses different values for each pixel according to the local area informationSkew Correction: Skew Correction Skew Correction methods are used to align the paper document with the coordinate system of the scanner. Main approaches for skew detection include correlation, projection profiles.Slant Removal : Slant Removal The slant of handwritten texts varies from user to user. Slant removal methods are used to normalize the all characters to a standard form.Slant Removal : Slant Removal Entropy The dominant slope of the character is found from the slope corrected characters which gives the minimum entropy of a vertical projection histogram. The vertical histogram projection is calculated for a range of angles ± R. In our case R=60, seems to cover all writing styles. The slope of the character, ,is found from: The character is then corrected by using:Feature Extraction: Feature Extraction In feature extraction stage each character is represented as a feature vector, which becomes its identity. The major goal of feature extraction is to extract a set of features, which maximizes the recognition rate with the least amount of elements. Due to the nature of handwriting with its high degree of variability and imprecision obtaining these features, is a difficult task. Feature extraction methods are based on 3 types of features: Statistical features Structural features Global transformations and moments features1.Statistical Features: 1.Statistical Features Representation of a character image by statistical distribution of points takes care of style variations to some extent. The major statistical features used for character representation are: Zoning Projections and profilesZoning: Zoning The character image is divided into NxM zones. From each zone features are extracted to form the feature vector. The goal of zoning is to obtain the local characteristics instead of global characteristicsZoning – Density Features: Zoning – Density Features The number of foreground pixels, or the normalized number of foreground pixels, in each cell is considered a feature. Darker squares indicate higher density of zone pixels.Zoning – Direction Features : Zoning – Direction Features Based on the contour of the character image For each zone the contour is followed and a directional histogram is obtained by analyzing the adjacent pixels in a 3x3 neighborhoodZoning – Direction Features : Zoning – Direction Features Based on the skeleton of the character image Distinguish individual line segments Labeling line segment information Line segments are coded with a direction number 2 = vertical line segment 3 = right diagonal line segment 4 = horizontal line segment 5 = left diagonal line segment Formation of feature vector through zoning number of horizontal lines number of right diagonal lines number of vertical lines number of left diagonal lines number of intersection pointsProjection Histograms: Projection Histograms The basic idea behind using projections is that character images, which are 2-D signals, can be represented as 1-D signal. These features, although independent to noise and deformation, depend on rotation. Projection histograms count the number of pixels in each column and row of a character image. Projection histograms can separate characters such as “m” and “n” .Profiles: Profiles The profile counts the number of pixels (distance) between the bounding box of the character image and the edge of the character. The profiles describe well the external shapes of characters and allow to distinguish between a great number of letters, such as “p” and “q”.2.Structural Features: 2.Structural Features Characters can be represented by structural features with high tolerance to distortions and style variations. This type of representation may also encode some knowledge about the structure of the object or may provide some knowledge as to what sort of components make up that object. Structural features are based on topological and geometrical properties of the character, such as aspect ratio , cross points, loops, branch points, horizontal curves at top or bottom, etc.Structural Features: Structural FeaturesStructural Features: Structural Features A structural feature extraction method for recognizing handwritten characters Two types of features: Horizontal projection histograms Vertical projection histograms3.Global Transformations - Moments: 3.Global Transformations - Moments The Fourier Transform (FT) of the contour of the image is calculated. Since the first n coefficients of the FT can be used in order to reconstruct the contour, then these n coefficients are considered to be a n -dimesional feature vector that represents the character. Central moments that make the process of recognizing an object scale, translation, and rotation invariant . T he original image can be completely reconstructed from the moment coefficients.Classification: Classification ANN is used for classification purpose.Results so far…: Results so far…References: References [1] S. Suresh, S.N. Omkar, and V. Mani, " Parallel Implementation of Back-Propagation Algorithm in Networks of Workstations " IEEE Transactions on Parallel and Distributed Systems, vol. 16, no. 1,pp.24-34, January 2005 [2] Wani, M.A.; Rashid, S., “ Parallel Algorithm for Control Pattern Recognition ” , Fourth International Conference on Machine Learning and Applications, 15-17 Dec. 2005 IEEE. [3] Andrew T. Wilson, " Off-line Handwriting Recognition Using Artificial Neural Networks ", University of Minnesota, Morris, wilsonat@mrs.umn.edu [4] Robert Sabourin, Luiz E. Soares de Oliveira, Edouard Lethelier, Fl´avio Bortolozzi, " A New Segmentation Approach for Handwritten Digits “ [5] Michael Blumenstein and Brijesh Verma, " A Neural Network for Real-World Postal Address Recognition ", School of Information Technology, Faculty of Engineering and Applied Science, Griffith University, Australia