#Google drive scan ocr for free#

The goal is achievable, in real time line, given the ingenuity and capability of the current Samskruth -Computer related human resources. This would call for ab initio rethinking on the use of ' Romanization approach and foundation for advancing the thought of 'Samskruth Programming language / Panini machine development'. I had raised this issue almost a decade ago on the need for advancing on the IPA diacritic model approach for Brahmi language scripts.

#Google drive scan ocr for free#

The indology model of ' script evolution and scripting conventions in multiple scripts of india ' is an approximation of about 45 to 60% ! in reading the manu-scripts.įurther deliberations on these points are clearly issues with commercial implications and technology oriented not appropriate for free for all publicity- public forum debate. The need for home team is to work abinitio with clarity on 'OCR for Samskruth, based on Brahmi language Native Design'- a feature which is not present in the current standards of Unicode Devanagari glyph model script shape construction. Contra, many of them are based on alien model of Roman script design. The current designs of OCR have not understood the ' Lipi-Shaastra' model of Devanagari Brahmi models.

How is this relevant to Samskruth- OCR ? When Venkat Veera Raghavan noted:, the challenge ahead is clear. Rest is the story in Kishkinda kaanda leading to Yuddha-Kanda !. This situation lead the team even to kill themselves using various means like jumping to sea, starve to death et al.Īt this time Jambavan is said to have stepped in and prompted the quiet Hanuman to undertake the work, using his dormant powers. That is the bar set for action is 200% plus.

The messenger needs to go, validate and return safe. So the message was clear ! One way 100% jump ability is NOT enough for the work. The Chief, Prince Angada said: he can jump hundred miles and land on other side, but was not sure of jumping back to bring the information about the correctness of Sita at Lanka ! This could not be an acceptable option, for it involved not only not getting the information validated but also risking their future king. The Vanara-sena TEAM started asking the question: Who amongst us is capable of jumping across the ocean and bring message about sita ?Įach team member carried out a self-assessment and declaration: I can jump one way - ten, fifty, seventy.ninety, ninety five. But the tools to achieve the goal was the missing element. Vanara-sena had the information: Sita is captive in the city 'Lanka' on the other end of the sea, hundred yojanas away. Vanara-sena had the directive from King Sugriva to see and bring information and where with all report on Sita and failure was marked with capital punishment ! It was death either way - jump in to ocean or return with failure in mission. The vanara-sena ( with angada the chief, the wise old veteran Jambavan, the young enthusiasts nala- neela et al, and quiet hanuman. Story (of Ramayana for Samskruth- OCR design) : Why ? The ( narrated) story (from Ramayana) on ocean jumping may help to understand this demand for OCR quality for Samskruth: A language sensitive for a 'change ( add / drop/ modification) of even a single swara (?!)-Varna ( ?!) - Maatraa ( ?!). I would set the quality standard expectancy at 200% accuracy ! This also answers why 90 to 95% accuracy is unacceptable. If I am to ask the question differently : What quality standard of OCR is needed for Samskruth document ? if current OCR solutions are not satisfactory for what we want, what would be the call for Samskruth OCR ? for manuscripts which did not use computer-fonts ?!

I agree with the observation of Ajit Gargeswari.