Vehicle License Plate Detector in Compressed Domain

Abstract

Data compression techniques allow data size to be reduced prior to data transmission and involve decompression upon transfer. This study shows for the first time that license plate (LP) detection can be accomplished without full decompression of the encoded data. Therefore, by determining in advance which images are required for LP recognition, computational costs of the system can be reduced. The proposed approach is realized on High Efficiency Video Coding (HEVC) based compressed video sequences. Two methods are provided that generate images from HEVC attributes. Fully decoded pixel domain images are also generated for comparative purposes from the same encoded data. The YOLO V3 Tiny Object Detector is used in order to detect LPs in the generated images. EnglishLP, a public dataset, is used to interpret the findings in terms of speed and precision and for comparison with previous studies. An additional contribution of the paper is that a new compressed domain LP database has been created and made publicly available, comprising images captured by a commercial license plate recognition system. Using at least two-orders-of-magnitude less amount of data, the proposed compressed domain LP detector achieved similar precision and recall values to those of the state-of-the-art LP detection schemes tested on both datasets. Moreover, the proposed method results in more than 30% saving in inference time. The results suggest that the proposed method can be utilized for rapid video archive searching applications.

Towards Automatic License Plate Detection

“…In [ 16 ], a YOLO V3 tiny object detector is introduced to detect license plates on images of the high efficiency video coding domain. This work reports a new compressed domain license plate database, which comprises images that are captured by a commercial license plate recognition system.…”

Compressed domain classification of remote sensing scene images based on sub-band data fusion

“…Savcı et al 12 utilizes only macroblock type and corresponding macroblock addresses information from the Advanced Video Coding (AVC) video to detect fire in the video. Beratoglu et al 13 save 30% processing time for the license plate recognition. They take High Efficiency Video Coding (HEVC) based compressed video sequences as input and detect license plate without full decompression.…”

“…Similar to that work, Beratoglu et al 14 use the block partitioning structure of the HEVC standard to find the license plate location of the given video. Using region information and motion vectors of the given video streams, which are compressed with HEVC format to find anomalies worked by C ¸avaş et al 13 Most of the compressed domain image analysis is based on wavelet transforms. Töreyin et al 15 proposed a smoke detection algorithm based on Markov model and wavelet transform for the videos that are compressed with MJPEG2000 and captured with a fixed camera.…”

Optimising Faster R-CNN Training to Enable Video Camera Compression for Assisted and Automated Driving Systems

“…The authors demonstrated a benefit in terms of the decreased bandwidth requirements for the same level of inference accuracy when comparing compressed data to lossless. Furthermore, they discuss a decrease of the NN inference time, which can be key in low-latency time sensitive automotive applications [12]. Tanaka et al carried out a similar investigation: compression based on HEVC and tracking based on the combination of YOLO V3 and SORT models [13].…”

“…in excess of 40 Gb/s) for higher levels of autonomy (levels 3-5), current wired vehicle data networks are inadequate to reliably transmit the required data amount [3]- [5]. With the increased demand of automotive cameras providing high resolution (8)(9)(10)(11)(12) and the required high dynamic range (HDR) to cope with the luminosity variations when driving (e.g. bright sun in front of the sensor when travelling in a dark tunnel), cameras can significantly contribute to the amount of generated data by the sensor suite; moreover multiple cameras are required to provide 360°coverage of a vehicle's surroundings.…”