GATR (Globally-scalable Automated Target Recognition) is a Lockheed Martin software system for real-time object detection and classification in satellite imagery on a worldwide basis. GATR uses GPU-accelerated deep learning software to quickly search large geographic regions. On a single GPU it processes imagery at a rate of over 16 square km/sec (or more than 10 Mpixels/sec), and it requires only two hours to search the entire state of Pennsylvania for gas fracking wells. The search time scales linearly with the geographic area, and the processing rate scales linearly with the number of GPUs. GATR has a modular, cloud-based architecture that uses the Maxar GBDX platform and provides an ATR analytic as a service. Applications include broad area search, watch boxes for monitoring ports and airfields, and site characterization. ATR is performed by deep learning models including RetinaNet and Faster R-CNN. Results are presented for the detection of aircraft and fracking wells and show that the recalls exceed 90% even in geographic regions never seen before. GATR is extensible to new targets, such as cars and ships, and it also handles radar and infrared imagery.
WoodYOLO: A Novel Object Detector for Wood Species Detection in Microscopic Images
Fraunhofer ITWM（弗劳恩霍夫信息技术与微系统研究所）；Thünen Institute of Wood Research（滕木材研究所）；Institute of Machine Learning and Analysis (IMLA)（机器学习与分析研究所）；Offenburg University（奥芬堡大学）
Automated Model Design and Benchmarking of 3D Deep Learning Models for COVID-19 Detection with Chest CT Scans
Hong Kong Baptist University（香港浸会大学）；Hong Kong University of Science and Technology（香港科技大学）；Hangzhou Dianzi University（杭州电子科技大学）；Tsinghua University（清华大学）
COVID-19疫情已在全球蔓延数月。由于其传播性和高致病性严重威胁着人们的生命，因此准确快速地检测COVID-19感染至关重要。最近的许多研究表明，基于深度学习（DL）的解决方案可以帮助基于胸部CT扫描检测COVID-19。然而，大多数现有的工作集中在2D数据集，这可能会导致低质量的模型，因为真正的CT扫描是3D图像。此外，报告的结果跨越了不同数据集的广泛范围，比较相对不公平。在本文中，我们首先使用三种最先进的3D模型（ResNet 3D 101，DenseNet 3D 121和MC 318）在三个公开的胸部CT扫描数据集上建立基线性能。然后，我们提出了一个可区分的神经结构搜索（DNAS）框架，自动搜索的3D DL模型的3D胸部CT扫描分类与Gumbel Softmax技术，以提高搜索效率。我们进一步利用类激活映射（CAM）技术对我们的模型提供的结果的可解释性。实验结果表明，我们的自动搜索模型（CovidNet 3D）在三个数据集上的性能优于基线人类设计的模型，模型大小小几十倍，准确性更高。此外，结果还验证了CAM可以很好地应用于CovidNet 3D中的COVID-19数据集，为医疗诊断提供可解释性。
Synthetic Data-Driven Multi-Architecture Framework for Automated Polyp Segmentation Through Integrated Detection and Mask Generation
Department of Advanced Computing, Morgan State University（高级计算系，摩根州立大学）；Department of Computer and Electrical Engineering, Morgan State University（计算机与电气工程系，摩根州立大学）
