Search

A high-quality dataset is a basic requirement to ensure the training quality and prediction accuracy of a deep learning network model (DLNM). To explore the influence of label image accuracy on the performance of a concrete crack segmentation network model in a semantic segmentation dataset, this study uses three labelling strategies, namely pixel-level fine labelling, outer contour widening labelling and topological structure widening labelling, respectively, to generate crack label images and construct three sets of crack semantic segmentation datasets with different accuracy. Four semantic segmentation network models (SSNMs), U-Net, High-Resolution Net (HRNet)V2, Pyramid Scene Parsing Network (PSPNet) and DeepLabV3+, were used for learning and training. The results show that the datasets constructed from the crack label images with pix-el-level fine labelling are more conducive to improving the accuracy of the network model for crack image segmentation. The U-Net had the best performance among the four SSNMs. The Mean Intersection over Union (MIoU), Mean Pixel Accuracy (MPA) and Accuracy reached 85.47%, 90.86% and 98.66%, respectively. The average difference between the quantized width of the crack image segmentation obtained by U-Net and the real crack width was 0.734 pixels, the maximum difference was 1.997 pixels, and the minimum difference was 0.141 pixels. Therefore, to improve the segmentation accuracy of crack images, the pixel-level fine labelling strategy and U-Net are the best choices.

Wheat is the third most important cereal crop in Nepal after rice and maize both in area and production, but its productivity of 2.3 tonne ha - 1 is very less compared to other developed countries (6 tonne ha -1 for Switzerland and China) in the world. The main cause of low wheat yield in Nepal is the improper and inadequate use of fertilizer devoid of site specific nutrient management practices. Therefore, a farmers' field experiment was conducted during November 2015 to April 2016 to rectify the best fertilizer management options at two sites of Damak and Gauradaha in Jhapa district in eastern-Terai of Nepal using Nutrient Expert®-Wheat model. The research was accomplished in Randomized Complete Block Design with 2 treatments and 20 replications, considering farmers' field as replication. Two treatments included in the experimentation were NE® (Nutrient Expert Recommendation) and FFP (Farmer's Fertilizer Practices). The statistical result revealed the highly significant difference in terms of number of effective tiller m -2 , plant height, filled grain per spike, spike length, grain, straw and biological yields and harvest index. The highest yield (4.71 tonne ha -1 ) was obtained from NE field than FFP (2.99 tonne ha -1 ). On an average, NE based practices produced 58 % higher yield in comparison to FFP. NE based treatment produced significantly higher biomass yield, yield attributes and cost-benefit ratio than FFP treatments. Field experiment validation confirmed that the Nutrient Expert® Wheat model could be used as the most adoptable and practical precision decision support system tool to make a more authentic fertilizer recommendation in eastern-Terai of Nepal., (© 2020 The Author(s).)