Deep Homography for License Plate Detection

Abstract

The orientation of plate images in license plate recognition is one of the factors that influence its accuracy. In particular, tilted plate images are harder to detect and recognize characters with than aligned ones. To this end, the rectification of plates in a preprocessing step is essential to improve their performance. We propose deep models to estimate four-corner coordinates of tilted plates. Since the predicted corners can then be used to rectify plate images, they can help improve plate recognition in plate recognition. The main contributions of this work are a set of open-structured hybrid networks to predict corner positions and a novel loss function that combines pixel-wise differences with position-wise errors, producing performance improvements. Regarding experiments using proprietary plate images, one of the proposed modes produces a 3.1% improvement over the established warping method.

Unsupervised Domain Adaptive Corner Detection in Vehicle Plate Images

“…and MNet. are from the existing works [6,7]. Those models work in a regression way, predicting normalized coordinates of the corners.…”

“…Corner Detection: Finding the four corner positions of vehicle plates for the rectification purpose can be considered as a keypoint detection problem. CNN-based methods [6,7] use features or latent vectors extracted through a sequence of convolution layers to regress corner coordinates. Instead of regressing corners, the methods [8,9] tried to directly generate rectified images.…”

Is Vehicle Plate Corner Prediction by Vision Transformer Better than CNNs?

“…Methods [5,6] using CNN for corner prediction use features extracted from convolutional layers or predict corner coordinates from a latent representation created by an autoencoder. While these methods fnd the coordinates required for rectifcation through warping, a method for directly creating a rectifed image [7] has also been proposed.…”