This work draws attention to the large fraction of near-duplicates in the training and test sets of datasets widely adopted in License Plate Recognition (LPR) research. These duplicates refer to images that, although different, show the same license plate. Our experiments, conducted on the two most popular datasets in the field, show a substantial decrease in recognition rate when six well-known models are trained and tested under fair splits, that is, in the absence of duplicates in the training and test sets. Moreover, in one of the datasets, the ranking of models changed considerably when they were trained and tested under duplicate-free splits. These findings suggest that such duplicates have significantly biased the evaluation and development of deep learning-based models for LPR. The list of near-duplicates we have found and proposals for fair splits are publicly available for further research at https://raysonlaroca.github.io/supp/lpr-train-on-test/

翻译：本研究揭示了车牌识别（LPR）研究中广泛采用的数据集内训练集与测试集中存在大量近重复样本的现象。这些重复样本指代不同图像但显示相同车牌的实例。我们在该领域两个最常用的数据集上开展的实验表明，当采用公平划分（即训练集与测试集无重复样本）对六种主流模型进行训练与测试时，识别率显著下降。此外，在其中某个数据集上，当模型在无重复划分条件下进行训练与测试时，模型排名发生了显著变化。这些发现表明，此类重复样本已对基于深度学习的LPR模型评估与开发产生显著偏差。我们发现的近重复样本列表及公平划分方案已在https://raysonlaroca.github.io/supp/lpr-train-on-test/ 公开，供后续研究使用。