GEOSPATIAL EXPLOITATION PRODUCTS Big Geospatial Data + Deep Learning + High Performance Computing = Geospatial Intelligence Bingcai Zhang Tech Fellow GXP Xplorer and SOCET GXP are registered trademarks of BAE Systems. All other brands, product names, and trademarks are property of their respective owners. This document gives only a general description of the product(s) or service(s) offered by BAE Systems and, except where expressly provided otherwise, shall not form part of any contract. From time to time, changes may be made in the products or conditions of supply. Approved for public release on 02/25/2016. 1
Do We Have Enough Parking? Demo 300 drone images GSD = 3.5cm 1600 cars detected 99% detection accuracy 6 pixel positional accuracy 10 degree orientation accuracy 0.001% false positive error rate Data from Prof. Dunn Model trained with 7.5cm GSD 2
What Is Deep Learning? It works just like the brain (least favorite definition according to LeCun) car 0.5 car 1.0 not car 0.5 not car 0.0 IEEE SPECTRUM 3
Simplicity Learning Object detection very complex Breakup objects Learn one type of object at a time Detect one type of object at a time Inspired by 4 year old pre-school best learning practice Learn one alphabet per week Learn letter A five days in a row (reinforcement learning) Inspired by drug discovery One specific drug for one specific disease No panacea Based on two decades of research experience Transform a complex problem into its simplest components Solve each component one at a time 4
Data Normalization vs. Data Augmentation Scale normalization vs. scale augmentation Color normalization vs. color augmentation Rotation normalization (geospatial images) 5
Simplicity Learning vs. Non-Simplicity Learning vehicle and anything else vehicle, stop sign, and anything else 6
Handcrafted Automatic Feature Extraction 7
3D Features (3D Glasses) 8
Rotation Variant Object Detection 30 20 10 0-88 -80-72 -64-56 -48-40 -32-24 -16-8 0 8 16 24 32 40 48 56 64 72 80 88 avg min max std -10-20 -30 9
Rotation Variant Object Detection 2 1.2 probability 1 0.8 0.6 probability 0.4 0.2 0-88 -80-72 -64-56 -48-40 -32-24 -16-8 0 8 16 24 32 40 48 56 64 72 80 88 10
Rotation Variant Object Detection 3 11
Rotation Variant Object Detection 4 12
Singular Classification soft max : c e j 1 zi e z j Open-ended negative training examples problem Not work just like the brain Three new algorithms reducing false positive by 10 times 13
Human vs. Machine Human: Radoslav Gaidadjiev Two master s degree Twenty years of experience with imagery Machine: DeepObject with one K40 GPU Human achieved accuracy of 99.9% Understand parking lot Machine achieved accuracy of 99.3% Demo Everyone participates human vs. machine 14
Quality Training Samples = $ Quality training samples are the new currency in deep learning Non-deep learning: AOD failed to recognize a car and an image analyst found this missing car A DR is generated and the cost to fix this DR is $1000 Deep learning: This missing car could be automatically collected as a positive training example and added to the training sample database Train deep learning network again with the new training example database Cost could be as low as $10 Potential cost saving is very significant 15
Mistakes = Quality Training Samples We learn from our mistakes (so does deep learning) Not all training samples are created equal Mistakes are more likely to have greater gradient Have stronger influence on decision boundary Quality of training samples is as important as quantity of training samples Data augmentation increases quantity Mistakes increases quality Future geospatial intelligence software should collect users intelligence Every mistake could translate into an enhancement to geospatial intelligence software 16
Future Intelligent Geospatial Intelligence System An intelligent system that could become smarter and smarter by learning from its mistakes An intelligent system that could detect and monitor defense relevant objects at 99% accuracy With 99% accuracy, this may be the game changer in geospatial intelligence domain Significantly reduce software engineering and enhancement costs 17
Questions? Dr. Bingcai Zhang 858-592-5218 bingcai.zhang@baesystems.com 18