PatternNet: A Benchmark Dataset for Performance Evaluation of Remote Sensing Image Retrieval

TL;DR

This paper introduces PatternNet, a large-scale, purpose-built dataset for remote sensing image retrieval, addressing limitations of existing datasets and enabling more effective development and evaluation of retrieval methods.

Contribution

The paper presents PatternNet, a new extensive dataset specifically designed for RSIR, along with a comprehensive review and baseline evaluation of over 35 retrieval methods.

Findings

PatternNet contains 38 classes with 800 images each.

Extensive baseline results for 35+ RSIR methods are provided.

The dataset enables improved development of deep learning-based RSIR techniques.

Abstract

Remote sensing image retrieval(RSIR), which aims to efficiently retrieve data of interest from large collections of remote sensing data, is a fundamental task in remote sensing. Over the past several decades, there has been significant effort to extract powerful feature representations for this task since the retrieval performance depends on the representative strength of the features. Benchmark datasets are also critical for developing, evaluating, and comparing RSIR approaches. Current benchmark datasets are deficient in that 1) they were originally collected for land use/land cover classification and not image retrieval, 2) they are relatively small in terms of the number of classes as well the number of sample images per class, and 3) the retrieval performance has saturated. These limitations have severely restricted the development of novel feature representations for RSIR, particularly the recent deep-learning based features which require large amounts of training data. We therefore present in this paper, a new large-scale remote sensing dataset termed "PatternNet" that was collected specifically for RSIR. PatternNet was collected from high-resolution imagery and contains 38 classes with 800 images per class. We also provide a thorough review of RSIR approaches ranging from traditional handcrafted feature based methods to recent deep learning based ones. We evaluate over 35 methods to establish extensive baseline results for future RSIR research using the PatternNet benchmark.