Delphi: Read text from screenshot images (OCR)
Budget: $30 – $250 USD
Your task is to use Delphi 12 (or compatible) to create a class that will take an input of a filename that is an image in BMP format and output all the texts from that image as a String. In other words, the main functionality will be:
Function ReadTextsFromImage(const Filename : String) : String;
There shall also be a simple demo application that allows me to test this code by giving this function a filename and it will output the texts from the image.
This task must be implemented purely in Delphi code, without the use of any third party libraries, DLL files, applications or anything like that. The Delphi code will open the image, analyze its content to read the texts via OCR that you implement. You can use third party libraries if these libraries are open source with BSD style license, and made with Delphi code, and you approve the use of such code with me before you start.
The input images shall be screenshots. In other words, the character recognition system does not need to handle things like rotated letters, handwriting or anything like that.
Attached are five sample screenshots. Your implementation must be able to read 100% of these texts. The screenshots can originate from other computer system than the current user system. In other words, you cannot assume the screenshots will only use fonts from the current system. That being said, the system only needs to support fonts that are in these five sample screenshots.
In case of characters that look visually identical to each other, such as upper case i and lower case L, mistakes are allowed.
Lastly, you can assume the texts contain only characters from the Extended ASCII character set.
To avoid bidding spam and automated bids, please start your bid description text with the sentence "OCR guru here". I will ignore all bids that do not start with this sentence.
Function ReadTextsFromImage(const Filename : String) : String;
There shall also be a simple demo application that allows me to test this code by giving this function a filename and it will output the texts from the image.
This task must be implemented purely in Delphi code, without the use of any third party libraries, DLL files, applications or anything like that. The Delphi code will open the image, analyze its content to read the texts via OCR that you implement. You can use third party libraries if these libraries are open source with BSD style license, and made with Delphi code, and you approve the use of such code with me before you start.
The input images shall be screenshots. In other words, the character recognition system does not need to handle things like rotated letters, handwriting or anything like that.
Attached are five sample screenshots. Your implementation must be able to read 100% of these texts. The screenshots can originate from other computer system than the current user system. In other words, you cannot assume the screenshots will only use fonts from the current system. That being said, the system only needs to support fonts that are in these five sample screenshots.
In case of characters that look visually identical to each other, such as upper case i and lower case L, mistakes are allowed.
Lastly, you can assume the texts contain only characters from the Extended ASCII character set.
To avoid bidding spam and automated bids, please start your bid description text with the sentence "OCR guru here". I will ignore all bids that do not start with this sentence.