No abstract

Huge number of images are acquired and analysed every day for a range of applications in civil infrastructure. One such application is the identification of cracks in concrete surface images, which is a challenge owing to their low contrast and resolution, blurriness, noise and information loss. Existing image enhancement algorithms improve either contrast or resolution to a rather limited extent. This paper proposes a Hybrid Image Enhancement (HIE) algorithm to improve both the contrast and resolution of concrete surface images using the Wavelet transform and Singular Value Decomposition (SVM). The enhanced concrete surface crack images are classified into specific crack types. The classification comprises preprocessing, crack detection, feature extraction and crack classification. The images are initially preprocessed using the Wiener filter to remove blurriness, following which cracks are detected using morphological operations and discontinuities in the segmented crack regions eliminated using the K-Dimensional Tree algorithm. Features are extracted from the segmented regions using statistical and geometric features. The image is classified thereafter into specific crack types using algorithms from three different neural network, kernel and tree based categories. The proposed HIE algorithm is validated using quantitative metrics and the results obtained are compared with those from State-of-the-Art methods and datasets. The results have shown that the HIE algorithm offers significantly improved accuracy of between 6% and 10% in the classification of concrete surface images.

Recent technological developments and improvement in the medical domain demands advancement, to address the issue of early disease detection. Also, the current pandemic has resulted in considerable progress of improvement in the medical domain, through online consultation by physicians for different diseases using clinical reports and medical images. A similar process is adopted in developing a Visual Question Answering (VQA) system in the medical field. In this paper, existing medical VQA datasets, appropriate techniques, suitable quantitative metrics, real time challenges and, the implementation of one VQA approach with algorithms and performance evaluation are discussed. The medical VQA datasets collected from multiple sources are represented in different perspectives (organwise, planewise, modality-type and abnormality-type) for a better understanding and visualization. Then the techniques used in VQA are subsequently grouped and explained, based on evolution, complexity in the dataset and the need for semantics in understanding the questions. In addition, the implementation of a VQA approach using VGGNet and LSTM is carried out for existing and improved datasets, and analyzed with accuracy and BLEU score metrics. The improved datasets, created through dataset reduction and augmentation approaches, resulted in better performance than the existing datasets. Finally, the challenges of the medical VQA domain are examined in terms of datasets, combining techniques, and modifying the parameters of existing performance metrics for future research.

No abstract