• Yechterla Monika
• Dr S. Jhansi Rani

The use of computers in many engineering fields is widely accepted. The field of geotechnical engineering is not far behind. The use of computers not only automates the process of soil characterization, but also makes it more objective. The chances of human error are minimized, thus saving a lot of energy, time and most importantly, money. The present work proposes a new technique to develop a correlation between soil image features and physical properties of soil materials through Digital Image Processing. Classification of soil is the dissolution to soil sets to Particular group having a like characteristics and similar manners. Almost all countries do product exporting, in which Those countries exporting higher agricultural product are very much depend on the soil characteristics. Thus, soil characteristics identification and classification is very much important. Identification of the soil type helps to avoid agricultural product quantity loss. A classification for engineering purpose should be based mainly on mechanical properties. This paper explains support vector machine and CNN based classification of the soil types. Soil classification includes steps like image acquisition, image pre-processing, feature extraction and classification. The texture features of soil images are extracted using the low pass filter, Gabor filter and using colour quantization technique. Artificial neural networks (ANNs) have been widely used for the analysis of remotely sensed imagery. In particular, convolutional neural networks (CNNs) are gaining more and more attention in this field. CNNs have proved to be very effective in areas such as image recognition and classification, especially for the classification of large sets composed by two-dimensional images. The developed techniques include CNN is used to classify soils based on various detectable features such as soil moisture content, soil nutrients, soil structure, soil quality, soil pH, and soil texture. A huge amount of literature is available for soil classification methods using the machine learning methods

Soil classification, Machine Learning, ANN, SVM, CNN, HSV; Gabor filter

"A new deep convolutional neural network technique For image based soil classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 6, page no.639-644, June 2019, Available :http://www.jetir.org/papers/JETIR1907395.pdf

"A new deep convolutional neural network technique For image based soil classification", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 6, page no. pp639-644, June 2019, Available at : http://www.jetir.org/papers/JETIR1907395.pdf