Accelerating Hyperparameter Tuning of a Deep Learning Model for Remote Sensing Image Classification

Aach, Marcel; Sedona, Rocco; Lintermann, Andreas; Cavallaro, Gabriele; Neukirchen, Helmut; Riedel, Morris

doi:10.1109/igarss46834.2022.9883257

Cited by 2 publications

(2 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, with the availability of parallel computing environments makes hyperparameter tuning feasible and should not be only considered as future work by researchers. Mentioned tuning with minimal reporting 4 [39], [40] [41], [42] Mentioned tuning with reproducible reporting 6 [43], [41], [44]- [47] Although it is not exhaustive, our short literature survey reflected that most of the work done in RS using DL does not discuss the importance of hyperparameter tuning. Since the cost of hyperparameter tuning is high due to the complexity of the DL model and the size of training data, it is essential to discuss the cost of tuning and also methods and means, such as parallel computing, that can be utilized for hyperparameter tuning.…”

Section: E Hyperparameter Tuning In DL Based Rsi Analysismentioning

confidence: 99%

“…In [47], the authors introduce changing a hyperparameter, specifically the batch size, to improve the training time for DL-based RS applications. In [47], the authors emphasize the use of hyperparameter tuning in DL and also employ Ray Tune [23] with a restricted set of parameters to explore the hyperparameter space.…”

Section: ) Hyperparameter Tuning: Done and Reportedmentioning

confidence: 99%

See 1 more Smart Citation

Hyperparameter Optimization for Large-Scale Remote Sensing Image Analysis Tasks: A Case Study Based on Permafrost Landform Detection Using Deep Learning

Perera,

Witharana,

Manos

et al. 2024

IEEE Access

View full text Add to dashboard Cite

Climate change pressure on the Arctic permafrost is rising alarmingly, creating a decisive need to produce Pan-Arctic scale permafrost landform and thaw disturbance information from remote sensing (RS) data. Very high spatial resolution (VHSR) satellite images can be utilized to detect ice-wedge polygons (IWPs) -the most important and widespread landform in the Arctic tundra region -across the Arctic without compromising spatial details. Automated analysis of peta-byte scale VHSR imagery covering millions of square kilometers is a computationally challenging task. Traditional semantic segmentation requires the use of task specific feature extraction with conventional classification techniques. Semantic complexity of VHSR images coupled with landscape heterogeneity makes it difficult to use conventional classification approaches to produce Pan-Arctic scale geospatial products. This leads to adapting deep convolutional neural network (DLCNN) approaches that have excelled in computer vision (CV) applications. Transitioning domains from everyday image understanding to remote sensing image analysis is challenging. This study aims to systematically investigate two main obstacles confronted when adapting DLCNNs in largescale RS image analysis tasks; 1) the limited availability labeled data sets and 2) the prohibitive nature of hyperparameter tunning when designing DLCNNs that can capture the rich characteristics embedded in remotely-sensed images. With a case study on the production of the first pan-Arctic ice-wedge polygon map using thousands of VHSR images, we demonstrate the use of transfer learning and the impact of hyperparameter tuning with a 16% improvement of the Mean Average Precision (mAP).

show abstract