FIELD OF INVENTION
Embodiments of a present disclosure relate to a digitization, and more particularly to a system for digitizing documents and a method to operate the same.
BACKGROUND
Digitization is a process of converting information into a digital format which is further used by a computing system for numerous possible reasons. Digitization involves a plurality of tasks such as watching a feeder system, checking quality of scan, interpreting handwritten data, identification of documents, identifying the accuracy in the interpreted data and the like.
However, the conventional approach requires a lot of human involvement in performing the plurality of tasks which is costly and a time-consuming process. Moreover, due to human involvement, chances of human error and inefficiencies are very high.
Hence, there is a need for an improved system for digitizing documents and a method to operate the same in order to address the aforementioned issues.
BRIEF DESCRIPTION
In accordance with an embodiment, a method for digitizing documents is provided. The method includes extracting one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents. The method also includes comparing one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images. The method also includes identifying one or more parameters from at least one identified image by using an identification technique. The method also includes extracting one or more characters from one or more identified parameters using a character segmentation technique. The method also includes recognizing one or more extracted characters using a character recognition technique. The method also includes creating one or more words by mapping the one

or more words with a database using a mapping technique to convert in a digitized format.
[0006] In accordance with another embodiment of the disclosure, a system for
digitizing documents is disclosed. The system includes one or more processors. The system also includes a document format identification subsystem operable by the one or more processors. The document format identification subsystem is configured to extract one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents. The document format identification subsystem is also configured to compare one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images. The system also includes a parameter identification subsystem operable by the one or more processors. The parameter identification subsystem is configured to identify one or more parameters from at least one identified image by using an identification technique. The system also includes a character recognition subsystem operable by the one or more processors. The character recognition subsystem is configured to extract one or more characters from one or more identified parameters using a character segmentation technique. The character recognition subsystem is also configured to recognize one or more extracted characters using a character recognition technique. The system also includes a word creation subsystem operable by the one or more processors. The word creation subsystem is configured to create one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format.
[0007] To further clarify the advantages and features of the present disclosure, a
more particular description of the disclosure will follow by reference to specific embodiments thereof, which are illustrated in the appended figures. It is to be appreciated that these figures depict only typical embodiments of the disclosure and are therefore not to be considered limiting in scope. The disclosure will be described and explained with additional specificity and detail with the appended figures.
BRIEF DESCRIPTION OF THE DRAWINGS

The disclosure will be described and explained with additional specificity and detail with the accompanying figures in which:
[0008] FIG. 1 is a flow diagram representing steps involved in a method for
digitizing documents in accordance with an embodiment of the present disclosure;
[0009] FIG. 2 is a block diagram representation of a system for digitizing
documents in accordance with an embodiment of the present disclosure;
[0010] FIG. 3 is a block diagram of an exemplary embodiment of the system for
digitizing documents of FIG. 2 in accordance with an embodiment of the present disclosure; and
[0011] FIG. 4 is a block diagram representation of a general computer system in
accordance with an embodiment of the present disclosure.
[0012] Further, those skilled in the art will appreciate that elements in the figures
are illustrated for simplicity and may not have necessarily been drawn to scale. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the figures by conventional symbols, and the figures may show only those specific details that are pertinent to understanding the embodiments of the present disclosure so as not to obscure the figures with details that will be readily apparent to those skilled in the art having the benefit of the description herein.
DETAILED DESCRIPTION
[0013] For the purpose of promoting an understanding of the principles of the
disclosure, reference will now be made to the embodiment illustrated in the figures and specific language will be used to describe them. It will nevertheless be understood that no limitation of the scope of the disclosure is thereby intended. Such alterations and further modifications in the illustrated system, and such further applications of the principles of the disclosure as would normally occur to those skilled in the art are to be construed as being within the scope of the present disclosure.
[0014] The terms "comprises", "comprising", or any other variations thereof, are
intended to cover a non-exclusive inclusion, such that a process or method that

comprises a list of steps does not include only those steps but may include other steps not expressly listed or inherent to such a process or method. Similarly, one or more devices or sub-systems or elements or structures or components preceded by "comprises... a" does not, without more constraints, preclude the existence of other devices, sub-systems, elements, structures, components, additional devices, additional sub-systems, additional elements, additional structures or additional components. Appearances of the phrase "in an embodiment", "in another embodiment" and similar language throughout this specification may, but not necessarily do, all refer to the same embodiment.
[0015] Unless otherwise defined, all technical and scientific terms used herein
have the same meaning as commonly understood by those skilled in the art to which this disclosure belongs. The system, methods, and examples provided herein are only illustrative and not intended to be limiting.
[0016] In the following specification and the claims, reference will be made to a
number of terms, which shall be defined to have the following meanings. The singular forms “a”, “an”, and “the” include plural references unless the context clearly dictates otherwise.
[0017] Embodiments of the present disclosure relate to a system for digitizing
documents and a method to operate the same is disclosed. The system includes one or more processors. The system also includes a document format identification subsystem operable by the one or more processors. The document format identification subsystem is configured to extract one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents. The document format identification subsystem is also configured to compare one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images. The system also includes a parameter identification subsystem operable by the one or more processors. The parameter identification subsystem is configured to identify one or more parameters from at least one identified image by using an identification technique. The system also includes a character recognition subsystem

operable by the one or more processors. The character recognition subsystem is configured to extract one or more characters from one or more identified parameters using a character segmentation technique. The character recognition subsystem is also configured to recognize one or more extracted characters using a character recognition technique. The system also includes a word creation subsystem operable by the one or more processors. The word creation subsystem is configured to create one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format.
[0018] FIG. 1 is a flow diagram representing steps involved in a method (10) for
digitizing documents in accordance with an embodiment of the present disclosure. In one embodiment, the method (10) may include receiving, by an input receiving subsystem, one or more documents by a scanner via a document feeder for scanning the one or more documents. In one embodiment, receiving the one or more documents by the scanner may include receiving a handwritten document, an image, a printed document and the like. Further, in one embodiment, the method (10) may include activating, by a hardware chip, an application programming interface (API) for scanning one or more received documents, wherein the hardware chip may be housed inside the scanner. Furthermore, in some embodiment, the method (10) may include monitoring, by a monitoring subsystem, the scanner continuously in real time to determine load of the one or more received documents in the scanner using an internet of things (IoT) based remote monitoring subsystem.
[0019] In one embodiment, the method (10) may include generating, by the
monitoring subsystem, an API call when an error being occurred while scanning the one or more corresponding documents. In some embodiment, the method (10) may include sending, by the monitoring subsystem, one or more scanned images to a server upon successful completion of the scanning of the one or more documents.
[0020] Further, the method (10) includes extracting, by a document format
identification subsystem, one or more features from the one or more corresponding scanned images using a feature extraction technique, wherein the one or more corresponding scanned images is a representative of the one or more scanned documents in step 20. As used herein, the term “feature extraction” describes the

relevant shape information contained in a pattern so that the task of classifying the pattern is made easy by a formal procedure.
[0021] In one embodiment, extracting the one or more features may include
extracting one or more low level features. In such embodiment, extracting the one or more low level features may include extracting contrast, coarseness and the like of the one or more corresponding images. In one embodiment, extracting the one or more features from the one or more corresponding scanned images using the feature extraction technique may include extracting the one or more features from the one or more corresponding scanned images using one of a principle components analysis (PCA) technique, a linear discriminant analysis (LDA) technique, an independent component analysis (ICA) technique and the like.
[0022] Furthermore, the method (10) also includes comparing, by the document
format identification subsystem, one or more extracted features of the one or more corresponding scanned images with one or more prestored features associated with one or more corresponding prestored documents to identify format of the one or more corresponding documents in step 30.
[0023] In such embodiment, comparing the one or more extracted features of the
one or more corresponding scanned images with one or more prestored features associated with one or more corresponding prestored documents may include comparing the one or more extracted features of the one or more corresponding scanned images with one or more prestored features associated with one or more corresponding prestored documents by checking distance of the one or more extracted features of the one or more corresponding scanned images from the one or more prestored features associated with the one or more corresponding prestored documents, wherein the distance between the one or more extracted features and the one or more prestored features may be calculated by using a Euclidean distance.
[0024] Further, in some embodiment, the method (10) may include identifying,
by the document format identification subsystem, the smallest distance of the one or more extracted features of at least one scanned image from the one or more prestored features of at least one prestored document. Then, the method (10) may include comparing the smallest distance of the one or more extracted features with a

predefined threshold value. Further, the method (10) may include identifying the format of the at least one scanned image when the smallest distance of the one or more extracted features of the at least one image is less than the predefined threshold value.
[0025] In another embodiment, the method (10) may include generating, by an
error generation subsystem, an error when the document format identification subsystem identifies the smallest distance of the one or more extracted features of at least one image from the one or more prestored features of the at least two prestored documents. In such embodiment, generating the error may include generating the notification of document not found, some update has happened in the prestored document and the like.
[0026] In one embodiment, the method (10) may include generating, by the
documents format identification subsystem, a query for a user upon identifying an updated document in a database, wherein the query represents “you want me to add an updated document with another name?”. Furthermore, the method (10) may include storing, by an updating subsystem, the one or more updated documents in a database with a second name when an old version of the first file is present in the database upon receiving a positive response from the user. In a current context, the positive response means “yes’.
[0027] In some embodiment, the method (10) may include generating, by the
document format identification subsystem, a signal for the hardware chip when the required document is not present in the database.
[0028] In one specific embodiment, the method (10) may include refining, by an
image pre-processing subsystem, at least one identified image by performing a plurality of steps involved in a pre-processing of the image using a pre-processing technique. In such embodiment, the method (10) may include removing, by a skew correction, an unwanted background from the at least one identified scanned image to straighten the at least one identified image. Further, in one embodiment, the method (10) may also include identifying, by a de-slanting, an average angle of the characters in the at least one identified image for identifying exact position of cut in the characters. In such embodiment, identifying the average angle of the characters in the at least one identified image may include identifying the lines from the at least one

identified image and then inter-line spacing is identified by computing a horizontal projection histogram, wherein the histogram represents the tangent of the inter-line spacing.
[0029] In some embodiment, the method (10) may include enhancing, by a
contrast enhancement, the dark and bright images by adjusting the contrast very frequently using a contrast enhancement algorithm. In such embodiment, enhancing the dark and bright images by adjusting the contrast very frequently using the contrast enhancement algorithm may include enhancing the dark and bright images by adjusting the contrast very frequently using one of a histogram equalization, a threshold histogram equalization, a low complexity histogram modification algorithm and the like.
[0030] Further, in one embodiment, the method (10) may include reducing, by a
noise filtration, unwanted noise from the at least one identified image using a noise filtration technique. In such embodiment, reducing the unwanted noise from the at least one identified image using the noise filtration technique may include reducing the unwanted noise from the at least one identified image using one of a linear filter, an adaptive filter, a gaussian filter and the like, wherein the noise filtration technique depends on a type of noise.
[0031] Furthermore, the method (10) may also include reducing, by binarization
process, an information contained within the at least one identified image by converting a greyscale image to black and white using an Otsu thresholding, wherein the black color represents a text and the white color represents a non-text. As used herein, the term “Otsu thresholding algorithm’ refers to a variance-based thresholding, wherein the algorithm works by searching for the threshold that minimizes the weight within-class variance, or put another way maximizes the between-class variance. Suppose, the threshold value is T, so to binarize the image, pixels less than the value T are set to 0, while pixels >= T are set to 1.
[0032] Furthermore, the method (10) also includes identifying, by a parameter
identification subsystem, one or more parameters from at least one pre-processed image using a parameter identification technique in step 40. In one embodiment,

identifying the one or more parameters may include identifying an inter-line spacing and a word gap.
[0033] In one embodiment, identifying the one or more parameters from at least
one pre-processed image may include identifying line from at least one pre-processed image by computing a horizontal projection histogram. In such embodiment, identifying the line may include identifying the line by checking the frequency of the one or more texts in the at least one pre-processed image, wherein the one or more texts may be represented as a curve. Further, the method (10) may include identifying, by the parameter identification subsystem, the inter-line spacing using histogram, wherein the inter-line spacing may be represented as a straight line in the histogram.
[0034] Similarly, identifying the one or more parameters may include identifying
the word gap from at least one pre-processed image by computing a vertical projection histogram upon identifying the line. The method (10) also includes extracting, by a character recognition subsystem, one or more characters from one or more identified parameters using a character segmentation technique in step 50.
[0035] In one embodiment, extracting the one or more characters from the one or
more identified parameters using the character segmentation technique may include extracting the one or more characters from the one or more identified parameters using a convolutional neural network. As used herein, the term “convolutional neural network” refers to a powerful image processing, artificial intelligence that use deep learning to perform both generative and descriptive tasks, often using machine vison that includes image and video recognition, along with recommender systems and natural language processing.
[0036] Further, the method (10) may include extracting, by the character
recognition subsystem, one or more features from one or more words using 5 layers of convolution neural network. Furthermore, the method (10) may include labelling the one or more extracted features to determine where the word should be cut by identifying the distance between each character in the word. Further, the method (10) may include identifying the average angle of each of the character of the word by using the de-slanting process for cutting the one or more words into characters.

[0037] The method (10) also includes recognizing, by the character recognition
subsystem, one or more extracted characters using a character recognition technique in step 60. In one embodiment, recognizing the one or more extracted characters using the character recognition technique may include recognizing the one or more extracted characters using a long short-term memory (LSTM). As used herein, the term “LSTM’ refers to an artificial recurrent neural network (RNN) architecture used in the field of deep learning. Unlike standard feedforward neural networks, LSTM has feedback connections. It can not only process single data points such as images, but also entire sequences of data such as speech or video. Furthermore, the method (10) may include recognizing each of the one or more characters by using 11 layers of backpropagation.
[0038] The method (10) also includes creating, by a word creation subsystem, one
or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format in step 70. In one embodiment, creating the one or more words by mapping the one or more words with the database using the mapping technique may include creating the one or more words by mapping the one or more words with the database using a natural language processing.
[0039] In one embodiment, the method (10) may include providing, by a training
subsystem, training to the algorithm with the prestored data for identifying the character. Further, in some embodiment, the method (10) may include checking the distance of the word which needs to be created from the one or more prestored words in the database.
[0040] The method (10) may also include correcting, by the word creation
subsystem, the word with a high confidence value when the distance of the at least one prestored word is minimum from the word, wherein the minimum distance is computed by comparing the distance with a second predefined threshold value. If the distance is less than the second predefined threshold value, then the distance is minimum, otherwise the distance is maximum.
[0041] In another embodiment, the method (10) may include correcting, by the
word creation subsystem, the word with a low confidence value when the distance of the of the at least one prestored word is more from the word. In yet another embodiment, the method (10) may include displaying, by the word creation

subsystem, the same word with a low confidence value when the word is not present in the database.
[0042] Furthermore, in one specific embodiment, the method (10) may include
analysing, by a semantic analysis subsystem, relation between one or more created words to check meaning of a sentence using an analysis technique. In such embodiment, analysing the relation between the one or more created words to check meaning of a sentence using the analysis technique may include analysing the relation between the one or more created words to check meaning of a sentence using a deep learning, wherein the deep learning enables one or more models to understand sentence structure and semantics. Further, the model is built as a representation of the entire sentence based on how the words are arranged and interact with each other.
[0043] Further, in one embodiment, the method (10) may include generating, by
an error generation subsystem, an error when the one or more words are not connected properly to make a complete sentence. In some embodiment, the method (10) may include correcting, by the word creation subsystem, the sentence if any grammatical error is found in the sentence.
[0044] In one specific embodiment, the method (10) may include formatting, by
a formatting subsystem, a semi structured digitized image into a structured image by cropping each of one or more relevant details of the digitized image and make a new cropped image, wherein the semi structured digitized image is a representative of the digitized format of the at least one scanned image. Further, method (10) may include copying, by the formatting subsystem, each of the relevant fields from the semi-structured digitized image to the structured image and further send the cropped image to character recognition subsystem to digitize the text present in the cropped image.
[0045] Further, in one embodiment, the method (10) may include monitoring, by
the monitoring subsystem, one or more details associated with work load by a user via a user interface. In such embodiment, monitoring the one or more details may include monitoring timestamps, status, output associated with the identified document, digitized document and the like.
[0046] Further, the method (10) may include performing, by the monitoring
subsystem, a plurality of actions by the user upon monitoring the one or more details

via the user interface platform. In such embodiment, performing the plurality of actions may include performing one of an approving the document, rejecting the document, uploading the correct document, requesting for rescan the document and the like.
[0047] Furthermore, the method (10) may include uploading, by a data
transmission subsystem, the data from the digitized document into the enterprise back office systems via an inbuilt robotic process automation (RPA) tool upon approval of digitized documents, wherein the back-office systems refers to an accounting, a finance, an inventory, fulfilment, and the like.
[0048] FIG. 2 is a block diagram representation of a system (80) for digitizing
documents in accordance with an embodiment of the present disclosure. The system (80) includes one or more processors (90). In one embodiment, the system (80) may include an input receiving subsystem configured to receive one or more documents by a scanner via a document feeder for scanning the one or more documents. In one embodiment, the one or more documents may include, but not limited to, a handwritten document, an image, a printed document and the like.
[0049] In one embodiment, a hardware chip may be housed inside the scanner,
wherein the hardware chip may be configured to activate an application programming interface (API) for scanning one or more received documents. In some embodiment, the hardware chip may include a monitoring subsystem configured to monitor the scanner continuously in real time to determine load of the one or more received documents in the scanner using an internet of things (IoT) based remote monitoring subsystem. In one embodiment, the monitoring subsystem may be configured to again generate an API call when an error being occurred while scanning the one or more corresponding documents. The monitoring subsystem may also be configured to send one or more scanned images to a server upon successful completion of the scanning of the one or more documents.
[0050] Further, the system (80) also includes a document format identification
subsystem (100) operable by the one or more processors (90). The document format identification subsystem (100) is configured to extract one or more features from the one or more corresponding scanned images using a feature extraction technique,

wherein the one or more corresponding scanned images is a representative of the one or more scanned documents.
[0051] In one embodiment, the one or more features may include one or more low
level features. In such embodiment, the one or more low level features may include contrast, coarseness and the like of the one or more corresponding images. In one embodiment, the feature extraction technique may include, but not limited to, a principle components analysis (PCA) technique, a linear discriminant analysis (LDA) technique, an independent component analysis (ICA) technique and the like.
[0052] Further, the document format identification subsystem (100) may be
configured to compare one or more extracted features of the one or more corresponding scanned images with one or more prestored features associated with one or more corresponding prestored documents to identify format of the one or more corresponding documents. In one embodiment, the document format identification subsystem (100) may be configured to compare the one or more extracted features of the one or more corresponding scanned images with the one or more prestored features by checking distance of the one or more extracted features of the one or more corresponding scanned images from the one or more prestored features associated with the one or more corresponding prestored documents, wherein the distance between the one or more extracted features and the one or more prestored features may be calculated by using a Euclidean distance. As used herein, the term “Euclidean distance” refers to distance between two points in any number of dimensions - the square root of the sum of the squares of the differences between the respective coordinates in each of the dimensions.
[0053] Further, the document format identification subsystem (100) may be
configured to identify the smallest distance of the one or more extracted features of at least one scanned image from the one or more prestored features of at least one prestored document.
[0054] Further, the document format identification subsystem (100) may be
configured to compare identified smallest distance of the one or more extracted features with a predefined threshold value. The document format identification subsystem (100) may also be configured to identify the format of the at least one

scanned image when the smallest distance of the one or more extracted features of the at least one image is less than the predefined threshold value.
[0055] In another embodiment, the system (80) may include an error generation
subsystem configured to generate an error when the document format identification subsystem (100) identifies the smallest distance of the one or more extracted features of at least one image from the one or more prestored features of the at least two prestored documents. In such embodiment, the error may be a document not found, some update has happened in the prestored document and the like.
[0056] In one embodiment, the document format identification subsystem (100)
may be configured to generate a query for a user upon identifying an updated document in a database, wherein the query represents “you want me to add an updated document with another name?”. Further, the system (80) may include an updating subsystem configured to store the one or more updated documents in a database with some another name when an old version of the first file is present in the database upon receiving a positive response from the user. In a current context, the positive response means “yes’.
[0057] In some embodiment, the document format identification subsystem (100)
may also be configured to generate a signal for the hardware chip when the required document is not present in the database.
[0058] Further, in one embodiment, the system (80) may also include an image
pre-processing subsystem configured to refine at least one identified image by performing a plurality of steps involved in a pre-processing of the image using a pre-processing technique. In such embodiment, a first step from the plurality of steps may include a skew correction, wherein the skew correction removes an unwanted background from the at least one identified scanned image to straighten the at least one identified image.
[0059] Furthermore, a second step from the plurality of steps may include a de-
slanting process which identifies an average angle of the characters in the at least one identified image for identifying exact position of cut in the characters. In one embodiment, the de-slanting process identifies the lines from the at least one identified image and then inter-line spacing is identified by computing a horizontal projection

histogram, wherein the histogram represents the tangent of the inter-line spacing. Further, the average angle of the characters is identified based on the identified inter-line spacing.
[0060] In some embodiment, a third step from the plurality of steps may include
contrast enhancement which enhances the dark and bright images by adjusting the contrast very frequently using a contrast enhancement algorithm. In such embodiment, the contrast enhancement algorithm may include, but not limited to, histogram equalization, threshold histogram equalization, a low complexity histogram modification algorithm and the like.
[0061] Further, a fourth step from the plurality of steps may include a noise
filtration configured to reduce unwanted noise from the at least one identified image using a noise filtration technique. In one embodiment, the noise filtration technique may include, but not limited to, a linear filter, an adaptive filter, a gaussian filter and the like, wherein the noise filtration technique depends on a type of noise.
[0062] Further, a fifth step from the plurality of steps may include a binarization
which reduces an information contained within the at least one identified image by converting a greyscale image to black and white using Otsu thresholding, wherein the black color represents a text and the white color represents a non-text.
[0063] Further, the system (80) also includes a parameter identification subsystem
(110) operable by the one or more processors (90). The parameter identification subsystem (110) is configured to identify one or more parameters from at least one pre-processed image using a parameter identification technique. In one embodiment, the one or more parameters may include inter-line spacing and word gap.
[0064] In one embodiment, the parameter identification subsystem (110) may be
configured to identify line from at least one pre-processed image by computing a horizontal projection histogram. In such embodiment, the parameter identification subsystem (110) may be configured to identify the line by checking the frequency of the one or more texts in the at least one pre-processed image, wherein the one or more texts may be represented as a curve. Further, the parameter identification subsystem (110) may be configured to identify the inter-line spacing using histogram, wherein the inter-line spacing may be represented as a straight line in the histogram.

[0065] Similarly, the parameter identification subsystem (110) may be configured
to identify word gap from at least one pre-processed image by computing a vertical projection histogram upon identifying the line. Further, the system (80) includes a character recognition subsystem (120) operable by the one or more processors (90). The character recognition subsystem (120) is configured to extract one or more characters from one or more identified parameters using a character segmentation technique. In one embodiment, the character segmentation technique may include a deep learning technique. In such embodiment, the deep learning technique may include a convolutional neural network.
[0066] The character recognition subsystem (120) may be configured to extract
one or more features from one or more words using 5 layers of convolution neural network. Further, the one or more extracted features may be labelled to determine where the word should be cut by identifying the distance between each character in the word. Further, the average angle of each of the character of the word is identified by using the de-slanting process for cutting the one or more words into characters.
[0067] Further, the character recognition subsystem (120) is also configured to
recognize one or more extracted characters using a character recognition technique. In one embodiment, the character recognition technique may include a long short-term memory (LSTM). Further, the LSTM includes a backpropagation configured to recognize each of the one or more characters properly using 11 layers.
[0068] Further, the system (80) also includes a word creation subsystem (130)
operable by the one or more processors (90). The word creation subsystem (130) is configured to create one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format, wherein the database may include a local database, a remote database and a database containing words with cut. In one embodiment, the mapping technique may include a natural language processing.
[0069] In one embodiment, the system (80) may include a training subsystem
configured to train the algorithm with the prestored data for identifying the character. In one embodiment, the word creation subsystem (130) may be configured to check the distance of the word from the one or more prestored words in the database. The

word creation subsystem (130) may also be configured to correct the word with a high confidence value when the distance of the at least one prestored word is minimum from the word, wherein the minimum distance is computed by comparing the distance with a second predefined threshold value. If the distance is less than the second predefined threshold value, then the distance is minimum, otherwise the distance is maximum.
[0070] In another embodiment, the word creation subsystem (130) may be
configured to correct the word with a low confidence value when the distance of the of the at least one prestored word is more from the word. In yet another embodiment, the word creation subsystem (130) may be configured to display the same word with a low confidence value when the word is not present in the database.
[0071] Further, in one specific embodiment, the system (80) may include a
semantic analysis subsystem configured to analyse relation between one or more created words to check meaning of a sentence using an analysis technique. In such embodiment, the analysing technique may include a deep learning, wherein the deep learning enables one or more models to understand sentence structure and semantics. Further, the model is built as a representation of the entire sentence based on how the words are arranged and interact with each other.
[0072] Further, the error generation subsystem may be configured to generate an
error when the one or more words are not connected properly to make a complete sentence. In one embodiment, the semantic analysis subsystem may be configured to correct the sentence if any grammatical error is found in the sentence.
[0073] In one embodiment, the system (80) may include a formatting subsystem
configured to format a semi structured digitized image into a structured image by cropping each of the relevant details of the digitized image and make a new cropped image. Further, the formatting subsystem may be configured to copy each of the relevant fields from the semi-structured digitized image to the structured image and further send the cropped image to character recognition subsystem to digitize the text present in the cropped image.
[0074] In one specific embodiment, the monitoring subsystem may be configured
to monitor one or more details associated with work load by a user via a user interface.

In such embodiment, the one or more details may include, but not limited to, timestamps, status, output associated with the identified document, digitized document and the like. In one embodiment, the monitoring subsystem may be configured to perform a plurality of actions by the user upon monitoring the one or more details via the user interface platform. In such embodiment, the plurality of actions may include approve the document, reject the document, upload the correct document, edit or request for rescan the document and the like.
[0075] Further, the system (80) may include a data transmission subsystem
configured to upload the data from the digitized document into the enterprise back office systems via an inbuilt robotic process upon approval of digitized documents, wherein the back-office systems refers to accounting, finance, inventory, fulfilment, and the like.
[0076] FIG. 3 is a block diagram of an exemplary embodiment of the system (140)
for digitizing documents of FIG. 2 in accordance with an embodiment of the present disclosure. A plurality of documents (150) inputs into a scanner (170), by an input receiving subsystem (160), via a document feeder. Upon receiving the plurality of documents, the scanner (170) starts scanning each of the plurality of documents (150) one by one. Upon scanning, a chip present in the scanner (170) sends the plurality of scanned documents to a server, where the server identifies type of each of the scanned document, by a document format identification subsystem (180), by comparing one or more features of each of the scanned documents with the plurality of prestored features.
[0077] Upon identifying the type of the document, a pre-processing is performed,
by an image pre-processing subsystem (200), on the plurality of scanned documents for enhancing the quality of the plurality of scanned documents. Further, a line identification from each of the pre-processed image or the scanned document is done, by a parameter identification subsystem (210), to identify inter-line spacing and word identification from each of the pre-processed image is also done, by the parameter identification subsystem (210), to identify the word gap.
[0078] Further, the angles of each of the character are determined by using a de-
slanting process and distance between each character in the word is also identified.

Upon determining the angle and the distance, the one or more identified words are segmented, by the character recognition subsystem (220), into characters. Further, the characters are recognized by using a character recognition technique.
[0079] Furthermore, the one or more words are created, by a word creation
subsystem (230), by mapping the one or more words with a database using a mapping technique to convert in a digitized format. For example, the word is “hypotheses”. The word creation subsystem (230) starts searching the word “hypotheses” in the database and calculates the distance of the word “hypotheses with each of the similar word present in the database. Suppose the word “hypotheses” is written wrongly as “hypatheses”. Then the word creation subsystem (230) corrects the word “hypatheses” as “hypotheses” with a high confidence value upon identifying the distance of character “a’ and “o’ in the word hypotheses.
[0080] Furthermore, the document format identification subsystem (180), the
parameter identification subsystem (210), the character recognition subsystem (220) and the word creation subsystem (230) are substantially similar to a document format identification subsystem (100), a parameter identification subsystem (110), a character recognition subsystem (120) and a word creation subsystem (130) of FIG. 2.
[0081] FIG. 4 is a block diagram of a general computer system (240) in
accordance with an embodiment of the present disclosure. The computer system (240) includes processor(s) (90), and memory (250) coupled to the processor(s) (90) via a bus (260).
[0082] The processor(s) (90), as used herein, means any type of computational
circuit, such as, but not limited to, a microprocessor, a microcontroller, a complex instruction set computing microprocessor, a reduced instruction set computing microprocessor, a very long instruction word microprocessor, an explicitly parallel instruction computing microprocessor, a digital signal processor, or any other type of processing circuit, or a combination thereof.
[0083] The memory (250) includes a plurality of subsystems stored in the form
of executable program which instructs the processor (90) to perform the configuration of the system illustrated in FIG. 1. The memory (250) has following subsystems: a document format identification subsystem (100), a parameter identification subsystem

(110), a character recognition subsystem (120) and a word creation subsystem (130) of FIG. 2.
[0084] Computer memory elements may include any suitable memory device(s)
for storing data and executable program, such as read only memory, random access memory, erasable programmable read only memory, electrically erasable programmable read only memory, hard drive, removable media drive for handling memory cards and the like. Embodiments of the present subject matter may be implemented in conjunction with program subsystems, including functions, procedures, data structures, and application programs, for performing tasks, or defining abstract data types or low-level hardware contexts. Executable program stored on any of the above-mentioned storage media may be executable by the processor(s) (90).
[0085] The document format identification subsystem (100) instructs the
processor(s) (90) to extract one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents.
[0086] The document format identification subsystem (100) instructs the
processor(s) (90) to compare one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images.
[0087] The parameter identification subsystem (110) instructs the processor(s)
(90) to identify one or more parameters from at least one identified image by using an identification technique.
[0088] The character recognition subsystem (120) instructs the processor(s) (90)
to extract one or more characters from one or more identified parameters using a character segmentation technique.
[0089] The character recognition subsystem (120) instructs the processor(s) (90)
to recognize one or more extracted characters using a character recognition technique.

[0090] The word creation subsystem (130) instructs the processor(s) (90) to create
one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format.
[0091] Various embodiments of the present disclosure provide a technical solution
to the problem of digitizing documents. The present disclosure provides an efficient system to improve the accuracy of the digitized documents by checking and correcting each phase simultaneously which is involved in digitizing documents. Also, the present system has ability to read multiple languages which makes the system unique and highly efficient.
[0092] While specific language has been used to describe the disclosure, any
limitations arising on account of the same are not intended. As would be apparent to a person skilled in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein.
[0093] The figures and the foregoing description give examples of embodiments.
Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment. For example, order of processes described herein may be changed and are not limited to the manner described herein. Moreover, the actions of any flow diagram need not be implemented in the order shown; nor do all of the acts need to be necessarily performed. Also, those acts that are not dependant on other acts may be performed in parallel with the other acts. The scope of embodiments is by no means limited by these specific examples.

WE CLAIM:
1. A method (10) for digitizing documents comprising:
extracting, by a document format identification subsystem, one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents; (20)
comparing, by the document format identification subsystem, one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images; (30)
identifying, by a parameter identification subsystem, one or more parameters from at least one identified image by using an identification technique; (40)
extracting, by a character recognition subsystem, one or more characters from one or more identified parameters using a character segmentation technique; (50)
recognizing, by the character recognition subsystem, one or more extracted characters using a character recognition technique; and (60)
creating, by a word creation subsystem, one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format. (70)
2. The method (10) as claimed in claim 1, wherein identifying the one or more parameters from at least one identified image comprises identifying an inter-line spacing and a word gap from at least one identified image.
3. The method (10) as claimed in claim 1, further comprising refining, by an image pre-processing subsystem, at least one identified image by performing a plurality of steps involved in a pre-processing of the image using a pre-processing technique.

4. The method (10) as claimed in claim 1, further comprising analysing, by a semantic analysis subsystem, relation between one or more created words to check meaning of a sentence using an analysis technique.
5. The method (10) as claimed in claim 1, further comprising generating, by a report generation subsystem, an error report of the document upon receiving notification, wherein the notification is a representative of an error received from at least one of the semantic analysis subsystem, the document format identification subsystem and the word creation subsystem.
6. A system (80) for digitizing documents comprising:
one or more processors (90);
a document format identification subsystem (100) operable by the one or more processors (90), and configured to:
extract one or more features from one or more corresponding images using a feature extraction technique upon receiving the one or more images, wherein the one or more corresponding images is a representative of one or more scanned documents,
compare one or more extracted features of the one or more corresponding images with one or more prestored features associated with the one or more corresponding documents to identify format of the one or more corresponding images;
a parameter identification subsystem (110) operable by the one or more processors (90), and configured to identify one or more parameters from at least one identified image by using an identification technique;
a character recognition subsystem (120) operable by the one or more processors (90), and configured to:
extract one or more characters from one or more identified parameters using a character segmentation technique,

recognize one or more extracted characters using a character recognition technique; and
a word creation subsystem (130) operable by the one or more processors (90), and configured to create one or more words by mapping the one or more words with a database using a mapping technique to convert in a digitized format.
7. The system (80) as claimed in claim 6, wherein the one or more parameters comprises an inter-line spacing and a word gap from at least one identified image.
8. The system (80) as claimed in claim 6, further comprising an image pre-processing subsystem configured to refine at least one identified image by performing a plurality of steps involved in a pre-processing of the image using a pre-processing technique.
9. The system (80) as claimed in claim 6, further comprising a semantic analysis subsystem configured to analyse relation between one or more created words to check meaning of a sentence using an analysis technique.
10. The system (80) as claimed in claim 6, further comprising a report generation subsystem configured to generate an error report of the document upon receiving notification, wherein the notification is a representative of an error received from at least one of the semantic analysis subsystem, the document format identification subsystem and the word creation subsystem.