Hybrid Cnn-Transformer Network for Interactive Learning of Challenging Musculoskeletal Images

Bi, Lei; Buehner, Ulrich; Fu, Xiaohang; Williamson, Tom; Choong, Peter; Kim, Jin Man

doi:10.2139/ssrn.4535797

Cited by 1 publication

(1 citation statement)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Deep learning-based object recognition algorithms, such as convolutional neural networks (CNNs), have achieved stateof-the-art performance in object recognition tasks, and, more recently, models such as Vision Transformer (ViT) are also achieving state-of-the-art (SOTA) performance [8,9]. These deep learning-based object recognition algorithms are highly dependent on the environmental factors which affect the quality of the training data, so model performance may deteriorate due to insufficient training data, large amounts of noise, and the presence of unlearned These deep learning-based object recognition algorithms are highly dependent on the environmental factors which affect the quality of the training data, so model performance may deteriorate due to insufficient training data, large amounts of noise, and the presence of unlearned environmental factors [10,11]. Therefore, it is important to make the environmental factors and quality of training data and input data the same [12].…”

Section: Introductionmentioning

confidence: 99%

Contrast Enhancement-Based Preprocessing Process to Improve Deep Learning Object Task Performance and Results

Wang,

Kim,

Kim

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Excessive lighting or sunlight can make it difficult to judge visually. The same goes for cameras that function like the human eye. In the field of computer vision, object tasks have a significant impact on performance depending on how much object information is provided. Light presents difficulties in recognizing objects, and recognition is not easy in shadows or dark areas. In this paper, we propose a contrast enhancement-based preprocessing process to obtain improved results in object recognition tasks by solving problems that occur due to light or lighting conditions. The proposed preprocessing process involves the steps of extracting optimal values, generating optimal images, and evaluating quality and similarity, and it can be applied to the generation of training and input data. As a result of an experiment in which the preprocessing process was applied to an object task, the object task results for areas with shadows or low contrast were improved while the existing performance was maintained for datasets that require contrast enhancement technology.

show abstract