Learning Objectness from Sonar Images for Class-Independent Object Detection

Detecting novel objects without class information is not trivial, as it is difficult to generalize from a small training set. This is an interesting problem for underwater robotics, as modeling marine objects is inherently more difficult in sonar images, and training data might not be available apriori. Detection proposals algorithms can be used for this purpose but usually requires a large amount of output bounding boxes. In this paper we propose the use of a fully convolutional neural network that regresses an objectness value directly from a Forward-Looking sonar image. By ranking objectness, we can produce high recall (96 %) with only 100 proposals per image. In comparison, EdgeBoxes requires 5000 proposals to achieve a slightly better recall of 97 %, while Selective Search requires 2000 proposals to achieve 95 % recall. We also show that our method outperforms a template matching baseline by a considerable margin, and is able to generalize to completely new objects. We expect that this kind of technique can be used in the field to find lost objects under the sea.

[1]  Matias Valdenegro-Toro,et al.  Submerged marine debris detection with autonomous underwater vehicles , 2016, 2016 International Conference on Robotics and Automation for Humanitarian Applications (RAHA).

[2]  Forrest N. Iandola,et al.  SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <1MB model size , 2016, ArXiv.

[3]  C. Lawrence Zitnick,et al.  Edge Boxes: Locating Object Proposals from Edges , 2014, ECCV.

[4]  Bernt Schiele,et al.  How good are detection proposals, really? , 2014, BMVC.

[5]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Matias Valdenegro-Toro,et al.  End-to-end object detection and recognition in forward-looking sonar images with convolutional neural networks , 2016, 2016 IEEE/OES Autonomous Underwater Vehicles (AUV).

[7]  Koen E. A. van de Sande,et al.  Selective Search for Object Recognition , 2013, International Journal of Computer Vision.

[8]  Matias Valdenegro-Toro Objectness Scoring and Detection Proposals in Forward-Looking Sonar Images with Convolutional Neural Networks , 2016, ANNPR.

[9]  Matias Valdenegro-Toro Real-time convolutional networks for sonar image classification in low-power embedded systems , 2017, ESANN.

[10]  Joaquim Salvi,et al.  Automatic detection of underwater chain links using a forward-looking sonar , 2013, 2013 MTS/IEEE OCEANS - Bergen.

[11]  Bernt Schiele,et al.  What Makes for Effective Detection Proposals? , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Yan Pailhas,et al.  Cascade of Boosted Classifiers for Rapid Detection of Underwater Objects , 2010 .

[13]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[14]  Neelima Chavali,et al.  Object-Proposal Evaluation Protocol is ‘Gameable’ , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[15]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[16]  Derek Hoiem,et al.  Category Independent Object Proposals , 2010, ECCV.

[17]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[18]  Thomas Deselaers,et al.  Measuring the Objectness of Image Windows , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.