Automatic detection of impact craters on Al foils from the Stardust interstellar dust collector using convolutional neural networks

TL;DR

This paper presents a convolutional neural network approach to accurately detect tiny impact craters on aluminum foils from NASA's Stardust mission, aiding the analysis of interstellar dust samples.

Contribution

It introduces a CNN based on VGG16 architecture specifically designed for identifying microscopic impact craters in Stardust foil images, improving detection accuracy.

Findings

High sensitivity and specificity in crater detection

Effective identification of sub-micrometer impact craters

Potential to enhance analysis of Stardust samples

Abstract

NASA's Stardust mission utilized a sample collector composed of aerogel and aluminum foil to return cometary and interstellar particles to Earth. Analysis of the aluminum foil begins with locating craters produced by hypervelocity impacts of cometary and interstellar dust. Interstellar dust craters are typically less than one micrometer in size and are sparsely distributed, making them difficult to find. In this paper, we describe a convolutional neural network based on the VGG16 architecture that achieves high specificity and sensitivity in locating impact craters in the Stardust interstellar collector foils. We evaluate its implications for current and future analyses of Stardust samples.