ICTACT Journal on Image and Video Processing

Open Access Open Access  Restricted Access Subscription Access
Open Access Open Access Open Access  Restricted Access Restricted Access Subscription Access

CNN Transfer Learning for Detection, Counting and Segmentation of Coconut Palms from Satellite Images


Affiliations
1 Indian Institute of Remote Sensing, Indian Space Research Organisation - Dehradun, India
     

   Subscribe/Renew Journal


Several Free and Open Source (FOSS) tools use Neural Networks for detection of objects from images and videos captured from hand-held imaging devices. Satellite based Remote Sensing images offer wide area coverage and hold potential for detecting, counting and mapping manmade objects, trees, etc., but, have embedded geospatial information and often have more than three bands. Hence, the existing FOSS tools are not able to directly process Remote Sensing images for Computer Vision (CV) applications. This research aims to devise a methodology to adapt a FOSS CV tool, namely the TensorFlow Object Detection (TFOD) API, for detection, counting and segmentation of coconut palms from satellite images and ascertain if the technique can facilitate automated census of coconut palms. Dataset of coconut palm crowns was custom-created using multi-band images from World View-3 satellite. The images were pan-sharpened, cropped and labelled. SSDLite MobileNet V2 CNN, which was pre-trained on COCO dataset, was subjected to transfer learning using coconut data on Tesla K80x GPU. This re-trained CNN could successfully detect and count coconut palms with F-1 score more than 96 %. Histogram thresholds were used to segment and delineate each detected coconut palm crown with 87 % accuracy. Assessment of relative health status of coconut palms was mapped using the Normalised Difference Red-Edge Index derived from satellite images. This study demonstrated that TFOD API can indeed be adapted for object detection and segmentation from Remote Sensing images, albeit with some limitations.

Keywords

Computer Vision, Object Detection, Segmentation, TensorFlow Object Detection API, Satellite Image.
Subscription Login to verify subscription
User
Notifications
Font Size

  • R. Vargas, A. Mosavi and L. Ruiz, “Deep Learning: A Review”, Advances in Intelligent Systems and Computing, Vol. 5, No. 2, pp. 1-14, 2017.

  • Y. Li, H. Zhang, X. Xue, Y. Jiang and Q. Shen, “Deep Learning for Remote Sensing Image Classification: A Survey”, Data Mining and Knowledge Discovery, Vol. 8, No. 6, pp. 1-17, 2018.

  • I. Goodfellow, Y. Bengio and A. Courville, “Deep Learning”, MIT Press, 2016.

  • TensorFlow, “TFOD API-GitHub”, Available at https://github.com/tensorflow/models/tree/master/research/object_detection, Accessed at 2020.

  • K. Patel, “Custom Object Detection using TensorFlow from Scratch”, Available at: https://towardsdatascience.com/custom-object-detection-using-tensorflow-from-scratch-e61da2e10087, Accessed at 2020.

  • Evan, “TensorFlow-Object-Detection-on-the-Raspberry-Pi”, Available at: https://github.com/EdjeElectronics/TensorFlow-Object-Detection-on-the-Raspberry-Pi, Accessed at 2020.

  • R. Balsys, “TensorFlow Object Detection Merged with Grabscreen Tutorial Part 2”, Available at: https://pylessons.com/Tensorflow-object-detection-merged-grab-screen-faster/, Accessed at 2020.

  • S. Paul, “Vehicle Number Plate Detection”, Available at: https://github.com/sayakpaul/Vehicle-Number-Plate-Detection, Accessed at 2020.

  • V. Sodha, “TensorFlow Object Detection API Tutorial - Training and Evaluating Custom Object Detector”, Available at: https://becominghuman.ai/tensorflow-object-detection-api-tutorial-training-and-evaluating-custom-object-detector-ed2594afcf73, Accessed at 2019.

  • S. Obadja, “Math Operators Object Detection”, Available at: https://github.com/stevenobadja/math_object_detection, Accessed at 2020.

  • Evan, “TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10”, Available at: https://github.com/EdjeElectronics/TensorFlow-Object-Detection-API-Tutorial-Train-Multiple-Objects-Windows-10, Accessed at 2020.

  • Coconut Development Board, “Statistics - Area, Production, Productivity of Coconut”, Available at: https://www.coconutboard.gov.in/Statistics.aspx, Accessed at 2020.

  • V.G. Chandrasekharan, V.C. Vasanthkumar, P.V. Preethakumari, R.P. Viswam and E.S. Vinod, “Consolidated Report Concurrent Estimation of Coconut Production in Kerala 2012-13”, Available at: https://www.coconutboard.in/images/Survey/report-kerala-2012-13.pdf, Accessed at 2013.

  • Digital Globe, “World View-3”, Available at: https://dg-cms-uploads-production.s3.amazonaws.com/uploads/document/file/95/DG2017_WorldView-3_DS.pdf, Accessed at 2019.

  • Hands-On Tensor Board, “TensorFlow Dev Summit 2017”, Available at: https://www.youtube.com/watch?v=eBbEDRsCmv4, Accessed at 2019.

  • J. Hui, “mAP (Mean Average Precision) for Object Detection”, Available at: https://medium.com/@jonathan_hui/map-mean-average-precision-for-object-detection-45c121a31173, Accessed at 2019.

  • R. Padilla, “Metrics for Object Detection”, Available at: https://github.com/rafaelpadilla/Object-Detection-Metrics, Accessed at 2019.

  • N.D. Gholba, “Detection of Coconut Palms and Allied Species from High Resolution Satellite Images using Deep Learning Techniques”, Master Thesis, Department of Computer Science, Andhra University, pp. 1-120, 2019.

  • T.T. Lin, “labelImg”, Available at: https://github.com/tzutalin/labelImg, Accessed at 2019.

  • D. Tran, “Raccoon Detector Dataset”, Available: https://github.com/datitran/raccoon_dataset, Accessed at 2019.

  • L. Vladimirov, “Training Custom Object Detector — TensorFlow Object Detection API tutorial documentation,” TensorFlow, 2018. Available at: https://tensorflow-object-detection-api-tutorial.readthedocs.io/en/latest/training.html. Accessed: 27-Mar-2019.

  • Tensorflow, “Tensorflow Detection Model Zoo”, Available at: https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/detection_model_zoo.md, Accessed at 2019.

  • M. Hugi, “Tensorflow Numpy Image Reshape [Grayscale Images] - Stack Overflow”, Available at: https://stackoverflow.com/questions/51872412/tensorflow-numpy-image-reshape-grayscale-images, Accessed at 2019.

  • SciPy, “NumPy v1.17 Manual: numpy.vsplit”, Available at: https://docs.scipy.org/doc/numpy/reference/generated/numpy.vsplit.html, Accessed at 2020.

  • P. Sharma, “Computer Vision Tutorial: A Step-by-Step Tutorial on Image Segmentation Techniques (Part 1)”, Available at: https://www.analyticsvidhya.com/blog/2019/04/introduction-image-segmentation-techniques-python/, Accessed at 2019.

  • X. Liming and Z. Yanchao, “Automated Strawberry Grading System based on Image Processing”, Computers and Electronics in Agriculture, Vol. 71, No. 1, pp. 32-39, 2010.

  • A. Danusasmita, “Image Detection Project Finding Strawberries”, Available at: https://github.com/andridns/cv-strawberry/blob/master/strawberry.ipynb, Accessed at 2019.

  • Rochan, “Crop Image in Tensorflow Object Detection API and Display It”, Available at: https://stackoverflow.com/questions/51572429/crop-image-in-tensorflow-object-detection-api-and-display-it, Accessed at 2020.

  • P.S. Thenkabail, P. Teluguntla, M.K. Gumma and ed-Edge Spectral Information for Leaf Area Index Retrieval”, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, Vol. 11, No. 5, pp. 1482-1492, 2018.

  • S. Mallick, “Apply ColorMap for pseudocoloring in OpenCV (C++ / Python)”, Available at: https://www.learnopencv.com/applycolormap-for-pseudocoloring-in-opencv-c-python/, Accessed at 2020.


Abstract Views: 181

PDF Views: 1




  • CNN Transfer Learning for Detection, Counting and Segmentation of Coconut Palms from Satellite Images

Abstract Views: 181  |  PDF Views: 1

Authors

Niranjan D. Gholba
Indian Institute of Remote Sensing, Indian Space Research Organisation - Dehradun, India
Shefali Agrawal
Indian Institute of Remote Sensing, Indian Space Research Organisation - Dehradun, India
Arun Babu
Indian Institute of Remote Sensing, Indian Space Research Organisation - Dehradun, India

Abstract


Several Free and Open Source (FOSS) tools use Neural Networks for detection of objects from images and videos captured from hand-held imaging devices. Satellite based Remote Sensing images offer wide area coverage and hold potential for detecting, counting and mapping manmade objects, trees, etc., but, have embedded geospatial information and often have more than three bands. Hence, the existing FOSS tools are not able to directly process Remote Sensing images for Computer Vision (CV) applications. This research aims to devise a methodology to adapt a FOSS CV tool, namely the TensorFlow Object Detection (TFOD) API, for detection, counting and segmentation of coconut palms from satellite images and ascertain if the technique can facilitate automated census of coconut palms. Dataset of coconut palm crowns was custom-created using multi-band images from World View-3 satellite. The images were pan-sharpened, cropped and labelled. SSDLite MobileNet V2 CNN, which was pre-trained on COCO dataset, was subjected to transfer learning using coconut data on Tesla K80x GPU. This re-trained CNN could successfully detect and count coconut palms with F-1 score more than 96 %. Histogram thresholds were used to segment and delineate each detected coconut palm crown with 87 % accuracy. Assessment of relative health status of coconut palms was mapped using the Normalised Difference Red-Edge Index derived from satellite images. This study demonstrated that TFOD API can indeed be adapted for object detection and segmentation from Remote Sensing images, albeit with some limitations.

Keywords


Computer Vision, Object Detection, Segmentation, TensorFlow Object Detection API, Satellite Image.

References