Ability to manually adjust the artificial neural network
The first step in resolving plaque detection problems is to identify the plaque. In the first phase of LPR, a limiting box is produced that fits exactly where the license plate is located. This is the main condition for successful LPR. Therefore, LP diagnosis should be relatively accurate. In this article, we have used the opencv library as well as contour and edge finder to detect this bug.
With the arrival of the new license plate image, various filters are first applied to the image and the main location of the license plate is identified using the contour. At this stage, more image processing libraries have been used. Finally, using conditions such as the height of the numbers or the width of the main location of the license plate numbers, the numbers are separated and entered into the constructed model. Our nerves will be given in the third stage.
At this stage, with the CNN neural network, as well as various filters and activity functions suitable for this network, along with determining the correct values for the hyperparameters, the characters identified in the previous stage will be classified.
A tool for converting dictionary files aka glossaries. Mainly to help use our offline glossaries in any Open Source dictionary we like on any modern operating system / device.
A simple personal website powered by flask in python
Build REST APIs with Neo4j and Flask, as quickly as possible!
Convert HTML to Markdown-formatted text.
An anthology of a variety of tools for the Persian language in Python
A collection of notebooks with TensorFlow and the Keras API for various deep-learning and machine learning problems
Unofficial toolkit for Eitaa messenger.
A collection of ML related stuff including notebooks, codes and a curated list of various useful resources such as books and softwares. Almost everything mentioned here is free (as speech not free food) or open-source.