Robust Vehicle Number Plate Text Recognition and Data Analysis Using Tesseract Ocr

¹ CSE department, Sreenidhi University, Ghatkesar, Hyderabad
² Information Technology, Vignana Bharathi Institute of Technology, Hyderabad
³ Data consultant (Data governance, Data analytics, EPM Enterprise performance management, AI&ML), SONIKS Consulting LLC, USA
⁴ CSE(AI&ML), TKR college of Engineering and Technology
⁵ Sreenidhi University, Ghatkesar, Hyderabad

Abstract

To detect the vehicle number plate the system should understand the character and integers determined on vehicles. The proposed methodology holds three phases: pre-process, extraction of features and recognition of text. These phases include some operations like grey scale, adaptive threshold, morphological for extraction of characters and numbers from different quality of the images in pre-processing stage. By transforming the images to grey scale, this can remove the extraneous colours and extracts the appropriate values. In morphological removes the borders and removes the background noises. Adaptive thresholding deals with color around the number plate which can increase the contrast. The outcome was transmitted to feature extraction which helps to identify single characters and numbers which may differentiate among many similar letters and numbers.Text recognition last uses OCR techniques to convert the acquired characteristics into readable alphabetic letters. These methods depend on advanced tools like the Tesseract OCR & OpenCV libraries. Tesseract’s OCR features make sure accurate character recognition, and OpenCV improves image processing and computer vision features.