Applying Polynomial Learning for Soil Detection Based on Gabor Wavelet and Teager Kaiser Energy Operator Kamel H. Rahouma 1(&) and Rabab Hamed M. Aly 2 1 Electrical Engineering Department, Faculty of Engineering, Minia University, Minia, Egypt kamel_rahouma@yahoo.com 2 The Higher Institute for Management Technology and Information, Minia, Egypt Abstract. Soil detection is playing an important role in the environmental research. It helps the farmers to determine what kind of plants they can have. Also, it may help to mix plants in certain areas or farm new types. The main target of this paper is to classify the different types of soil. On the other hand, there are many researches which focus on the classification and detection pro- cess based on different applications of image processing and computer vision. The paper has two main goals. The first goal is to improve the extraction of soil features based on Gabor wavelet transform but followed by the Teager-Kaiser Operator. The second goal is to classify the types of soil based on group method data handling (polynomial neural networks). We applied these methods using different data sets of soil. Compared with previous work and research, we achieved accuracy limits of (98%–100%) while the previous algorithms were accurate to the limits of (95.1%–98.8%). Behind this improvement in accuracy, there are the methods we used here including the Teager Kaiser operator with Gabor wavelet and polynomial neural networks which have been proved to be more accurate than the methods used before. Keywords: Soil detection  Gabor wavelet  Polynomial neural network (PNN)  Teager-Kaiser 1 Introduction Nowadays, machine learning and computer vision play an important role in environ- ment image analysis, especially in the detection of features processing (Bhattacharya and Solomatine 2006). Furthermore, the effective role of computer vision and image processing is to classify the type of different soils. The researchers try to make the classification very easier using the modern research technologies. Researchers try to improve the methods of extraction or classification. On the other hand, some researches have introduced some other methods for the features detection of soils or diseases of leaves of trees. In the following, we will show some of the recent researches and how the authors try to improve methods to classify and detect the medical images and environments images such as soils (Bhattacharya and Solomatine 2006; Lu et al. 2018; Odgers and McBratney 2018; Pham et al. 2017). © Springer Nature Switzerland AG 2020 A. E. Hassanien et al. (Eds.): AMLTA 2019, AISC 921, pp. 771–783, 2020. https://doi.org/10.1007/978-3-030-14118-9_75