Search

In the last decades, scene text recognition has gained worldwide attention from both the academic community and actual users due to its importance in a wide range of applications. Despite achievements in optical character recognition, scene text recognition remains challenging due to inherent problems such as distortions or irregular layout. Most of the existing approaches mainly leverage recurrence or convolution-based neural networks. However, while recurrent neural networks (RNNs) usually suffer from slow training speed due to sequential computation and encounter problems as vanishing gradient or bottleneck, CNN endures a trade-off between complexity and performance. In this paper, we introduce SAFL, a self-attention-based neural network model with the focal loss for scene text recognition, to overcome the limitation of the existing approaches. The use of focal loss instead of negative log-likelihood helps the model focus more on low-frequency samples training. Moreover, to deal with the distortions and irregular texts, we exploit Spatial TransformerNetwork (STN) to rectify text before passing to the recognition network. We perform experiments to compare the performance of the proposed model with seven benchmarks. The numerical results show that our model achieves the best performance., Comment: Accepted to ICMLA 2020

Introduction and Importance: Several authors have also made reference to a less prevalent condition known in elbow as ulnar nerve subluxation. However, this particular condition tends to manifest primarily in young individuals who engage in professional sports or activities involving extensive use of the forearm. A more severe form of ulnar nerve subluxation, which is ulnar nerve dislocation, gives rise to a characteristic dislocation and relocation of the nerve at the elbow during flexion and extension of the forearm. Due to the rarity of this condition in clinical settings and its predominant occurrence as subluxation in younger patients, there are instances where traumatic ulnar nerve dislocation can be overlooked and misdiagnosed with two commonly encountered pathological conditions as ulnar nerve entrapment or medial epicondylitis., Case Presentation: The authors present a 51-year-old male with chronic pain when moving his right forearm following a fall that caused a direct force injury to his elbow. The patient was misdiagnosed and treated as medial epicondylitis and early-stage ulnar nerve entrapment. However, the symptoms did not improve for a long time. The authors performed the ulnar nerve anterior transposition surgery using the subcutaneous transposition technique and the result is very good without any pain., Clinical Discussion: The ulnar nerve can naturally be subluxed or dislocated if Osborne's ligament is loose or when there are anatomical variations in the medial epicondyle. In some case, this ligament can be ruptured by trauma. The symptoms of ulnar instability are caused by friction neuritis. Dynamic ultrasound of the ulnar nerve in two positions show clearly this condition., Conclusion: Post-traumatic ulnar nerve dislocation is a rare condition, and the recurrent characteristic of it leads to neuritis or neuropathy. The condition can be overlooked or misdiagnosed as medial epicondylitis or early-stage ulnar nerve entrapment. The nerve transposition surgery will give good result., Competing Interests: We declare that we have no known competing financial interests or personal relationships with anyone that could have appeared to influence the work reported in this paper.Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article., (Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc.)