Sand fineness modulus prediction in construction sector using convolutional neural network: Fid-Bau Portal

Sand fineness modulus prediction in construction sector using convolutional neural network

Fahad, AL / Nayem, Naymul Hasan / Hossain, Md. Nashib / Rabbani, Md. Liton / Opu, Raihan Khan / Al Shuaeb, S M Abdullah

The construction industry relies heavily on the sand. In construction, the fineness modulus of sand is an important parameter. It impacts the relative proportions in the mix, the workability, the economy, the porosity, and the strength of the concrete. This standard specifies that sand ‘s fineness modulus should not be less than 2.3 and not more than 3.1. Sand’s fineness modulus refers to the mean size of its particles. There is remarkable success in predicting various fruits images, grains, vegetables, and soils using convolutional neural networks. Convolutional neural networks (CNN) are introduced here as a deep learning-based approach to sand fineness modulus value prediction. The CNN algorithm extracts automatic features, so this research was conducted on the latest CNN architecture, ResNet-50. There are many different types of artificial neural networks (ANNs), each with its own strengths and weaknesses. We chose to use a convolutional neural network (CNN) because CNNs are well suited for image classification tasks. CNNs are able to learn features from images automatically, which makes them less prone to overfitting than other types of ANNs and its gives more accuracy in result. Currently, the sand (FM) is calculated based on laboratory sieve analysis, which is accurate but a time-consuming process. For that, the instance method is necessary to determine the sand’s fineness modulus. We have proposed a novel image-based model to predict sand FM values. However, sand the fineness modulus (FM) can be quickly determined using images, but with low accuracy. In experiments using our proposed method, we achieved 94.6% accuracy. There is also evidence that the proposed image-based system performs better on each of the five standard assessment metrics, including accuracy, precision, recall, specificity, and the F-score, when predicting fineness modulus (FM) values.