Method and apparatus for transformation of dot text in an image into stroked characters based on dot pitches

A non-transitory computer readable storage medium that provides instructions, which when executed by a processor, cause said processor to perform operations comprising:

• determining a statistical neighborhood of a set of dots of an image, wherein dots are extracted from the image as the set of dots, wherein the statistical neighborhood includes a set of points and each point is associated with a position and a statistical measure accumulated at the position based on a relative position of dots with respect to neighbor dots from the set of dots of the image, and based on whether one or more dots that satisfy a shape and a size criteria are located at that position;
• computing a Fast Fourier Transform (FFT) across the set of points of the statistical neighborhood;
• and
• determining, based on the FFT of the set of points, a first orientation and a first distance between adjacent dots of characters along the first orientation, and a second orientation and a second distance between adjacent dots of the characters along the second orientation, wherein the first orientation, the first distance, the second orientation and the second distance are to be used to perform accurate optical character recognition.

A system for determining orientation and dot pitch of characters in an image, wherein dots are extracted from the image as a set of dots, the system comprising:

• one or more processors;
• and
• a non-transitory computer readable storage medium that stores code which when executed by the one or more processors, causes the system to:
• determine a statistical neighborhood of the set of dots of the image, wherein the statistical neighborhood includes a set of points and each point is associated with a position and a statistical measure accumulated at the position based on a relative position of dots with respect to neighbor dots from the set of dots of the image, and based on whether one or more dots that satisfy a shape and a size criteria are located at that position,
• compute a Fast Fourier Transform (FFT) across the set of points of the statistical neighborhood, and
• determine, based on the FFT of the set of points, a first orientation and a first distance between adjacent dots of characters along the first orientation, and a second orientation and a second distance between adjacent dots of the characters along the second orientation, wherein the first orientation, the first distance, the second orientation and the second distance are to be used to perform accurate optical character recognition.