A novel feature based algorithm for soil type classification

Uddin, Md. Romij; Hassan, Md. Rakib

doi:10.1007/s40747-022-00682-0

Cited by 18 publications

(4 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In 2022 Uddin, M. and Hassan, M., [21] presented a feature-based algorithm for soil detection. Frequent ϕ-Pixels, gradients oriented toward the quartile histogram, and feature selection are essential components for soil classification.…”

Section: Literature Surveymentioning

confidence: 99%

TEA SOIL: Self-Power IoT for Precision Soil Farming of Tea Plant Cultivation

2024

Preprint

View full text Add to dashboard Cite

In agriculture identifying the "nature of the soil" is a significant phase for serval sorts of landscapes. Even though numerous studies have been conducted for continuous soil health monitoring, however, existing techniques face challenges like labor-intensive, cost efficiency, need extensive field sampling and not suitable for tracking of soil parameters. To overcome these issues this paper proposed a novel IoT-based TeaSoil for precision soil monitoring of tea plant cultivation. The proposed TeaSoil’s end nodes are known as Tea Soil Monitoring Units (TSMU), are solar-powered and placed up for extended periods of time. The TSMU wirelessly sends soil NPK sensor, soil temperature, moisture, pH, organic matter, zinc, copper, iron, magnesium, manganese, carbon dioxide (CO2), and geo-location data using SigFox communication. SigFox receives the data and uploads it to the cloud server for data storage and analysis over a long period of time. On this receiver side, the soil classification is carried out using a deep learning-based EfficientNet for tea plant cultivation. A TSMU dashboard allows users to examine gathered data. The proposed Tea soil achieves an overall accuracy of 99.15%. The accuracy of proposed techniques attains 0.75%, 2.16%, and 20.32% better than Fast R-BAG, CNN, and CSMO. It is demonstrated that the proposed TeaSoil is suitable for precision agriculture based on the experimental results.

show abstract

Section: Literature Surveymentioning

confidence: 99%

TEA SOIL: Self-Power IoT for Precision Soil Farming of Tea Plant Cultivation

2024

Preprint

View full text Add to dashboard Cite

show abstract

“…The discussed models have improved classification accuracy and helped to reduce the cost of classification. Uddin and Hassan 24 suggested a computer vision-based method for an automated soil classification system to predict soil types. The algorithm combined quartile histogram-oriented gradients, the most frequent φ-pixels, and a new feature selection method.…”

Section: Literature Surveymentioning

confidence: 99%

Soil classification with multi-temporal hyperspectral imagery using spectral unmixing and fusion

Kaba,

Leloglu

2023

J. Appl. Rem. Sens.

View full text Add to dashboard Cite

.Soil maps are essential sources for a diverse range of agricultural and environmental studies; hence, the detection of soil properties using remote sensing technology is a hot topic. Satellites carrying hyperspectral sensors provide possibilities for the estimation of soil properties. But, the main obstacle in soil classification with remote sensing methods is the vegetation that has a spectral signature that mixes with that of the soil. The objective of this study is to detect soil texture properties after eliminating the effects of vegetation using hyperspectral imaging data and reducing the noise by fusion. First, the endmembers common to all images and their abundances are determined. Then the endmembers are classified as stable ones (soil, rock, etc.) and unstable ones (green vegetation, dry vegetation, etc.). This method eliminates vegetation from the images with orthogonal subspace projection (OSP) and fuses multiple images with the weighted mean for a better signal-to-noise-ratio. Finally, the fused image is classified to obtain the soil maps. The method is tested on synthetic images and hyperion hyperspectral images of an area in Texas, United States. With three synthetic images, the individual classification results are 89.14%, 89.81%, and 93.79%. After OSP, the rates increase to 92.23%, 93.13%, and 95.38%, respectively, whereas it increases to 96.97% with fusion. With real images from the dates 22/06/2013, 25/09/2013, and 24/10/2013, the classification accuracies increase from 70.51%, 68.87%, and 63.18% to 71.96%, 71.78%, and 64.17%, respectively. Fusion provides a better improvement in classification with a 75.27% accuracy. The results for the analysis of the real images from 2016 yield similar improvements. The classification accuracies increase from 57.07%, 62.81%, and 63.80% to 58.99%, 63.93%, and 66.33%, respectively. Fusion also provides a better classification accuracy of 69.02% for this experiment. The results show that the method can improve the classification accuracy with the elimination of vegetation and with the fusion of multiple images. The approach is promising and can be applied to various other classification tasks.

show abstract

“…As a result of the CNN model, crucial details from the input photos can be collected. To categorize pictures, CNN uses these signals [11]. The automated feature extraction was shown in Fig.…”

Section: Automatic Features Extractionmentioning

confidence: 99%

Sand fineness modulus prediction in construction sector using convolutional neural network

Fahad,

Nayem,

Hossain

et al. 2023

Asian J Civ Eng

View full text Add to dashboard Cite

BackgroundThe construction industry relies heavily on the sand. In construction, the neness 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 speci es that sand's neness modulus should not be less than 2.3 and not more than 3.1. Sand's neness modulus refers to the mean size of its particles. Methods/Analysis/Findings:There is a 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 neness modulus value prediction. The CNN algorithm extracts automatic features, so this research was conducted on the latest CNN architecture, ResNet-50. Currently, the sand (FM) is calculated based on laboratory sieve analysis, which is an accurate but time-consuming process. For that, the instance method is necessary to determine the sand's neness modulus. We have proposed a novel image-based model to predict sand FM values. However, sand the neness modulus (FM) can be quickly determined using images, but with low accuracy. Applications/Improvements: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 ve standard assessment metrics, including accuracy, precision, recall, speci city, and the F-score, when predicting Fineness Modulus (FM) values.

show abstract

A novel feature based algorithm for soil type classification

Cited by 18 publications

References 60 publications

TEA SOIL: Self-Power IoT for Precision Soil Farming of Tea Plant Cultivation

TEA SOIL: Self-Power IoT for Precision Soil Farming of Tea Plant Cultivation

Soil classification with multi-temporal hyperspectral imagery using spectral unmixing and fusion

Sand fineness modulus prediction in construction sector using convolutional neural network

Contact Info

Product

Resources

About