Agriculture is an important component of every individual`s livelihood. During farming infestation of insect pests in crops are inevitable. Manual detection and identification of type is the most challenging process. The proposed architecture will pave a way to develop an automatic detection system to do the identification of pest present in videos. Through this system, the pest infestation can easily be identified and suitable management techniques can be applied early to improve the quality of crops. The proposed architecture incorporates the following techniques for effective detection insects in crops: key frame extraction by calculating a threshold from histogram difference of consecutive frames, best frame selection by finding PSNR and MSE value of key frames, filtering, color image segmentation through K – Means clustering segmentation, feature extraction through Neural Network techniques with a pre trained network VGG19 and classification by using multi class SVM classifier. By using this proposed algorithm, this system identifies five types of pests namely, tuta absoluta, fall armyworm, leaf hopper, epilachna beetle and corn borer with accuracy of 98.46%.

Key frame extraction, K-Means clustering Segmentation, Convolutional Neural Networks (CNN), VGG19, Support Vector Machine (SVM)

[1] Bashish, Dheeb Al, Braik, Malik and Bani-Ahmad, Sulieman, “Detection and Classification of Leaf Diseases using K-means-based Segmentation and Neural-networks-based Classification”, Information Technology Journal, Vol.10, pg. 267-275, 2011
[2] Gondal, Muhammad Danish, Khan, Yasir Niaz , “Early Pest Detection from Crop using Image Processing and Computational Intelligence”, FAST-NU Research Journal (FRJ), Vol.1, Issue. 1, pg. 59-68, 2015
[3] Premalatha C., “Automatic Smart Irrigation System Using IOT”, International Journal of Scientific Research in Computer Science and Engineering, Vol.7, Issue.1, pp.1-5, 2019
[4] Denan Xia, Peng Chen, Bing Wang, Jun Zhang and Chengjun Xie (2018), “Insect Detection and Classification Based on an Improved Convolutional Neural Network”, Sensors MDPI.
[5] Uma Rani, Amsini,“Pest Identification in Leaf Images using SVM Classifier”, International Journal of Computational Intelligence and Informatics, Vol. 6: Issue.. 1, pg. 30-41,2016
[6] Dang, Chinh, Radha, Hayder (2014), “RPCA-KFE: Key Frame Extraction for Video using Robust Principal Component Analysis”, IEEE Transactions on Image Processing, Vol. 24: Issue. 11, pg. 3742-3753,2016
[7] Sakshi kathuria, “A Survey on Security Provided by Multi-Clouds in Cloud Computing”, International Journal of Scientific Research in Network Security and Communication , Vol. 6, Isuue:3,pg. 23-27
[8] Ghatak, Sanjoy,“Key Frame Extraction using Threshold Technique”, International Journal of Engineering Applied Sciences and Technology, Vol. 1, Issue. 8, pg. 51-56, 2016.
[9] Sladojevic, Srdjan, Arsenovic, Marko, Anderla, Andras, Culibrk, Dubravko and Stefanovic, Darko, “Deep Neural Networks Based Recognition of Plant Diseases by Leaf Image Classification”, Hindawi Publishing Corporation Computational Intelligence and Neuroscience, Vol. 2016, pg. 1-11, 2016
[10] Sreelakshmi M, Padmanayana ,“Early Detection and Classification of Pests Using Image Processing”, International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering, Vol. 3, Issue. 1, pg. 239-242, 2015