Description:FORM 2 THE PATENT ACT, 1970 (39 of 1970) COMPLETE SPECIFICATION (See section 10, rule 13) TITLE: A SYSTEM AND METHOD FOR RESTORING TEXT LEGIBILITY OF FADED MANUSCRIPTS INVENTORS RANI, Shobha N, Indian citizen #836A, Bogadi North Cross, 9th Cross Mysuru, Karnataka- 570026 NAIR, Bipin BJ, Indian citizen Kallara Veedu, Menamkulam, Kazhakuttom Trivandrum, Kerala- 695582 PATI, Peeta Basa, Indian citizen Flat C, Bldg 27, 6th Main,Vinayakanagar B Block, Bangalore, Karnataka- 560017 APPLICANTS Amrita Vishwa Vidyapeetham #114, 7th Cross, Bogadi II Stage, Mysuru, Karnataka, India – 570026 THE FOLLOWING SPECIFICATION PARTICULARLY DESCRIBES THE INVENTION AND THE MANNER IN WHICH IT IS TO BE PERFORMED A SYSTEM AND METHOD FOR RESTORING TEXT LEGIBILITY OF FADED MANUSCRIPTS CROSS-REFERENCES TO RELATED APPLICATIONS None. FIELD OF THE INVENTION The present invention generally relates to image enhancementand more particularly relates toa system and method to restore text legibility of faded manuscripts. BACKGROUND OF THE RELATED ART Ancient manuscripts have a rich cultural legacy wherein text strokes play an important role in proper reading of their content. The strokes present in a character help disambiguate the character or word from many possibilities. Many times, due to fading of the stroke, the content present may get misread and a completely different interpretation of the knowledge may be arrived at. A proper restoration of these keystrokes is essential for proper legibility enhancement of a document manuscript. Any image enhancement method is based on the features from a certain type of document. Hence, there are no universal image enhancement procedures that work with all types of photographs. At present, document enhancement approaches are designed as thresholding or filter based image analysis models (Bannigidad et al. 2016) or deep learning based models (Anvari et al.2021) according to the state of the art of literature. The former methods, which use filters, rely on the selection of noise reduction parameters, while the latter, using deep learning, require a large amount of metadata and CPU capacity to process. For improving both the visual quality and the intelligibility of deteriorated Arabic and Latin handwritten writings, a deep learning architecture called Generative Adversarial Networks (GAN) (Jemniet al. 2022) has been used that necessitates a large amount of metadata for both enhancement and recognition. In one of the research, traditional approaches like morphological operation, contrast stretching, and median filtering have been used in combination with a pre- processing to eliminate the effect of non-uniform illumination and few age marks in historical manuscripts made of paper type and palm leaves (International Journal of Engineering Research and Technology. ISSN 0974-3154, Volume 14, Number 3 (2021), pp. 219-226). However, this approach is suitable for texts with minor or no degradations. Another research addresses non-uniform illumination, foxing effect, show through effect, yellowish marks in ancient manuscripts and palm leaf documents. But, this research excludes samples of faded text strokes from the palm leaf category (https://doi.org/10.1080/00051144.2022.2042462). The US patent US10853638B2 relates to a method and system aimed at word segmentation, spotting, and recognition in handwritten paper documents yet does not use pre-processing methods suitable for documents with extreme fading or degradations. Preservation strategies and technologies accessible for Potala palace palm leaf manuscripts (Sa etal. 2022) focus on storage methods and types of deterioration in palm leaf manuscripts. In (Sathik et al. 2021) and (Athisayamani et al. 2021), researchers have used deep learning to recognize non-degraded palm leaf manuscripts of Tamil palm leaf manuscripts. Furthermore, the study of a binarization technique based on the Whale Optimization Algorithm (WOA) for palm leaf manuscript restoration (Alexander etal. 2020)emphasizes the importance of legible palm leaf manuscripts wherein the approach is primarily based on textual data obtained using adaptive thresholding techniques, demonstrating its potential for restoring faded text pixels. There are other studies related to the establishment of transcriptoriums/benchmarkings (Kesiman et al. 2018; Kesiman et al. 2015) for palm leaf manuscripts, preservation methods (Saxena et al. 2021), recognition (Inkeaw et al. 2015), (Chamchong et al. 2011), (Singh et al. 2021), (Guruprasad et al. 2021), segmentation (Peng et al. 2016), (Valy et al 2016), and binarization (Chamchong et al. 2010), (BJ et al. 2021), (Paulus et al. 2021), and other topics. Models of above mentioned works are meant to work with usually deteriorated palm leaf images and do not necessarily involve the recovery of faded text. Hence, there has been a need in the art for a method and system that particularly tackles issues like recovering fading text, blur reduction, or marginal noise removal.In this regard, the method and system to restore text legibility according to the present invention substantially departs from the conventional concepts and designs of the prior art.The present invention focus on restoring text legibility of severely degraded palm leaf documents with faded text strokes and other forms of degradation based on HSV colour space, thresholding methods, and morphological operations. These and other advantages will be more readily understood by referring to the following detailed description disclosed hereinafter with reference to the accompanying drawing and which are generally applicable to other systems and methods for recovering text legibility to fulfill particular application illustrated hereinafter. SUMMARY OF THE INVENTION According to one embodiment of the present subject matter, a system for restoring text legibility of faded manuscript is disclosed. The system includes an input unit adapted to receive images of the faded manuscript form plurality of sources. In various embodiments, the input unit of the system captures images in RGB format in various resolutions. The system further includes a transformer unitthat is configured to perform contrast adjustment and saturation adjustment on the images received from the input unit. The transformer unit performs contrast and saturation adjustment to obtain a HSV color space. The transformer unit also detects and eliminates hole and margin sections to generate a noise free image output. The system includes a restoration unit adapted to compute a global faded text restoration threshold using HSV channels of the noise free image output received from the transformer unit to generate a faded text recovered image. The system includes a post-restoration unit adapted to perform adaptive thresholding on the faded text recovered image received from the restoration unit to generate a binary image to set a RGB color space after removing noise and distortions and retaining faded text strokes. Further, the system includes a display unit adapted to display the RGB image received from the post-restoration unit and to provide the enhanced image output. The system also includes a memory unit for storing the output images. In various embodiments, the transformer unitis further configured to perform contrast adjustment on the received RGB image W^0to obtain contrast enhanced image W^1. In various embodiments, the transformer unit performs saturation adjustment to convert W^1to HSV color space W_hsv^1for highlighting the channels h, s and v. In various embodiments, the transformer unit applies global thresholding to channel h to obtain the thresholded image h_b1. In various embodiments, the transformer unit generates the hole and margin detection mask h_b2 by erosion of h_b1by structuring element S_e. In various embodiments, the transformer unitde-noises hole and margin sections of original RGB image W^0to generate pre-processed image W^2of RGB color space. In various embodiments, the restoration unitis further configured to convert the pre-processed image W^2to HSV color spaceW_hsv^2. In various embodiments, the restoration unit extracts saturation channel W_s^2 from the HSV color space W_hsv^2.In various embodiments, the restoration unit performs statistical feature computation of W_s^2 to determine the faded text stroke recovery threshold t_f. In various embodiments, the restoration unit extracts green channel W_g^2 from RGB color space W_RGB^2. In various embodiments, the restoration unit applies the faded text stroke recovery threshold t_f to obtain a binary image W_gbi^2. In various embodiments, the restoration unit perform enhancement of channels W_r^2,W_g^2,W_b^2 of W_RGB^2to obtain faded text recovered image W^3 of RGB color space, whereinW_gbi^2is used as reference mask for the enhancement. In various embodiments, the post-restoration unit is further configured to extract W_g^3 of RGB color space W_RGB^3. In various embodiments, the post-restoration unit performs adaptive thresholding on W_g^3to obtain binary image W_gbi^3. In various embodiments, the post-restoration unit enhances the binary image W_gbi^3by applying morphological closing for dilation and erosion of W_gbi^3by Se. In various embodiments, the post-restoration unit generate enhanced image W_output^4. According to another embodiment of the present subject matter, a method for restoring text legibility for faded manuscript is disclosed. The method involves receiving images of the faded manuscript in an input unit in RGB format in various resolutions from a plurality of sources. The method includes generating a noise free output in a transformer unit by performing contrast and saturation adjustment on images received from the input unit to obtain a HSV color space, detect and eliminate hole and margin sections. The method includes generating a faded text recovered image in a restoration unit after computing a global faded text restoration threshold using HSV channels of the noise free image output received from the transformer unit. The method further includes generating a binary image to set a RGB color space in a post-restoration unit by performing adaptive thresholding on the faded text recovered image received from the restoration unit after removing noise and distortions and retaining faded text strokes. Next, the method involves displaying the RGB image received from the post-restoration unit in a display unit. In various embodiments, the step of generating the noise free output in the transformer unit involves performing contrast adjustment on the received RGB image W^0 to obtain contrast enhanced image W^1. In various embodiments, step of generating the noise free output involves performing saturation adjustment to convert W^1to HSV color space W_hsv^1for highlighting the channels h, s and v. In various embodiments, step of generating the noise free output involves applying global thresholding to channel hto obtain the thresholded image h_b1followed bygeneratingthe hole and margin detection mask h_b2by erosion of h_b1by structuring element S_e. In various embodiments, step of generating the noise free outputfurther involves de-noising hole and margin sections of original RGB image W^0 to generate pre-processed image W^2of RGB color space. In various embodiments, generating the faded text recovered image in the restoration unit involves converting the pre-processed image W^2to HSV color space W_hsv^2. Step for generating the faded text recovered image includes extracting saturation channel W_s^2 from the HSV color space W_hsv^2 followed by performing statistical feature computation of W_s^2 to determine the faded text stroke recovery threshold t_f. This is followed by extracting green channel W_g^2 from RGB color space W_RGB^2. Step for generating the faded text recovered image further includes applying the faded text stroke recovery threshold t_fto obtain a binary image W_gbi^2. Finally, performing enhancement of channels W_r^2,W_g^2,W_b^2 of W_RGB^2to obtain faded text recovered image W^3of RGB color space, wherein W_gbi^2is used as reference mask for the enhancement in step for generating the faded text recovered image. In various embodiments, generating the binary image to set the RGB color space in the post-restoration unit includes extracting W_g^3 of RGB color space W_RGB^3 followed by performing adaptive thresholding on W_g^3 to obtain binary image W_gbi^3. Step for generating the binary image to set the RGB color space further involves enhancing the binary image W_gbi^3by applying morphological closing for dilation and erosion of W_gbi^3by S_eand generation of enhanced imageW_output^4. This and other aspects are disclosed herein. BRIEF DESCRIPTION OF THE DRAWINGS The invention has other advantages and features which will be more readily apparent from the following detailed description of the invention and the appended claims, when taken in conjunction with the accompanying drawings, in which: FIG.1: represents a system for restoring text legibility of faded manuscripts. FIG. 2: represents a method for restoring text legibility for faded manuscript FIG. 3: shows contrast adjustment using image contrast transformation FIG. 4: represents pre-processing workflow of faded text stroke recovery method. FIG. 5: represents a workflow of faded text recovery threshold computation. FIG. 6: shows outcome of faded text recovery process for images in figure 3. FIG.7: shows outcome of post enhancement. FIG. 8: depicts performance meter for no-reference metrics (a), (b) & (c) FIG. 9A: depicts the individual AL, ACM, and GLV readings in relation to the dataset category C1 FIG. 9B: depicts the individual AL, ACM, and GLV readings in relation to the dataset category C2 FIG. 9C: depicts the individual AL, ACM, and GLV readings in relation to the dataset category C3. FIG. 10: depicts the results of the proposed algorithm for sample 1 of category of C1. FIG. 11: depicts the results of the proposed algorithm for sample 2 of category of C2. FIG. 12: depicts the results of the proposed algorithm for sample 5 of category of C3.   DETAILED DESCRIPTION OF THE EMBODIMENTS While the invention has been disclosed with reference to certain embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the scope of the invention. In addition, many modifications may be made to adapt to a particular situation or material to the teachings of the invention without departing from its scope. Throughout the specification and claims, the following terms take the meanings explicitly associated herein unless the context clearly dictates otherwise. The meaning of "a", "an", and "the" include plural references. The meaning of "in" includes "in" and "on." Referring to the drawings, like numbers indicate like parts throughout the views. Additionally, a reference to the singular includes a reference to the plural unless otherwise stated or inconsistent with the disclosure herein. The present subject matter describes a system and methods for restoring text legibility of faded manuscript. A block diagram of a system for restoring text legibility of faded manuscript is illustrated in FIG.1, according to one embodiment of the present subject matter. The system 100 may primarily include an input unit102, a transformer unit103, a restoration unit104, a post-restoration unit 105and a display unit107. The system may include a memory unit 106. The input unit 102 may be configured to receive images of the faded manuscripts 101 from a plurality of sources. In various embodiments, the input unit captures images of the faded manuscripts in RGB format in various resolutions. The transformer unit 103 may be configured to perform contrast and saturation adjustment on images received from the input unit 102to obtain a HSV color space, detect and eliminate hole and margin sections to generate a noise free image output. The restoration unit 104is configured to compute a global faded text restoration threshold using HSV channels of the noise free image output received from the transformer unit to generate a faded text recovered image. The post-restoration unit 105is configured to perform adaptive thresholding on the faded text recovered image received from the restoration unit to generate a binary image to set a RGB color space after removing noise and distortions and retaining faded text strokes. The display unit 107may beadapted to display the RGB image received from the post-restoration unit. In various aspects, a method 200for restoring text legibility for faded manuscript is disclosed. The method 200includes the step 201of receiving images of the faded manuscript in an input unit102 in RGB format in various resolutions from a plurality of sources. The method 200 includes step 202 of generating a noise free output in a transformer unit 103 by performing contrast and saturation adjustment on images received from the input unit102 for obtaining a HSV color space, detect and eliminate hole and margin sections. The method further includes step 203 of generating a faded text recovered image in a restoration unit 104 after computing a global faded text restoration threshold using HSV channels of the noise free image output received from the transformer unit 103. Generating a binary image to set a RGB color space in a post-restoration unit 105 by performing adaptive thresholding on the faded text recovered image received from the restoration unit 104 after removing noise and distortions and retaining faded text strokes is included in step 204. The method further includes displaying 205the RGB image received from the post-restoration unit in a display unit107. In various embodiments, the input unit 102 may be adapted to read manuscript images101 available in digital form W^0on the input unit or at a remote location. The input unit 102 may read the images W^0as a single file or in batch mode wherein all the files from the unit are taken in together. The image filesW^0may be in the form of PDF and TIFF format to support storage of multiple images in a single file. The image files W^0may be further processed by transferring to the transformer unit 103. In various embodiments, the transformer unit 103 may transform the image file W^0into a format that may be used for mid-level processing. In one embodiment, the transformer unit 103may apply contrastadjustment to the image filesW^0to obtainW^1. The transformer unit may apply contrast adjustment to improve the gradient details in order to aid in the recovery of faded text strokes. In various embodiments, the transformer unit may apply contrast adjustment f_1to RGB image W^0with (x_i,y_j) as a pixel position of W^0with i=1,2,3…m and j= 1,2,3…n. In one embodiment, applying f_1may produce contrast enhanced imageW^1. Function f1 applies a transformation ‘t’ on the input intensity level r^0 of W^0 (x_i,y_j) from i=1,2,3…m and j= 1,2,3…nproducing an adjusted intensity r^1 with respect to the gray levels lying in the minimum rangel_(min ) tol_maxofW0, may be computed as shown in (1). f_1 (t):W^0→W^1 andr^1=t(r^0)…..(1) wherein,gray level lies in l_(min )≤ r^0≤l_max and l_(min )≤ r^1≤l_max In various embodiments, the transformer unit 103may improve the contrast of micro image details difficult to capture with the naked eye wherein images with higher contrast levels typically show more intensity variation than images with lower contrast levels. The transformer unit may shift low contrast intensity levels towards considerably darker shades and may shift high contrast intensity levels towards brighter shades resulting in a change in the degree of intensity differentiation between foreground and background pixels. FIG. 3 shows the contrast adjustment using the transformer unit’s image contrast transformation wherein FIG. 3(a) is the original image and FIG. 3(c) is its intensity profile, FIG. 3(b) is the result of contrast adjustment and its intensity profile is represented in FIG. 3(d). In various embodiments, the transformer unit may be used to improve the contrast of micro image details that are difficult to capture with the naked eye. The transformer unit may convert the contrast enhanced images to HSV color spaceW_hsv^1 for highlighting the channels h, s and v.The transformer unit may further detect and eliminate hole and margin sections to create noise- free document image of the original image file. The noise-free document image may be further processed by transferring to the restoration module104. In various embodiments, the transformer unit 103may perform saturation adjustment to convert RGB image W^1 to HSV color space W_hsv^1. In various embodiments, the transformer unit may analyse the image details available in the HSV color space of image W^1 to detect the hole and margin regions.In various embodiments, HSV color space provides opportunity to deal with par visual effect, blurring and low brightness in RGB images as it separates the brightness of the image. In one embodiment, this conversion may highlight the chromaticity details in channel ‘h’ abbreviated as hue, saturation details in channel ‘s’ as saturation and brightness details in channel ‘v’ wherein the image details in channels ‘h’, ‘s’ and ‘v’ may not overlap with each other and effectively separates color and brightness. In various embodiments, this conversion may help in clear interpretation and understanding of specific color details pertaining to various sections of images. In various embodiments, hue values may vary from 0 to 360 degrees, with red (0-60), yellow (60-120), green (120-180), cyan (180-240), blue (240-300), and magenta (240-300) being the most common hues (300- 360). In various embodiments, saturation refers to the percentage of grey in an image that ranges from 0% to 100%. In one embodiment, when value ‘v’is set to ‘0’ the image appears completely black with no brightness, whereas bigger values increase brightness, resulting in colour discrimination, and vice versa. In various embodiments, channels‘s’ and ‘v’may be used to determine hole and margin regions. In one embodiment, the transformer unit103 may apply global thresholding f_2to channel ‘h’with gray level threshold of g_th, wherein given channels hue ‘h’, saturation ‘s’ and value ‘v’ of HSV color space W_hsv^1. In various embodiments, if l_r with r= l_min…l_min represents a gray level at h (x_i,y_j) for i=1,2,3…m and j=1,2,3…n, then a gray level l_r that is below g_th may be mapped to 0 and above g_th may be mapped to 1 as given in (2). f_(2 ) (h,g_th )=h→h_b1…..(2) such that l_r