Lossless and Near-Lossless Compression Algorithms for Remotely Sensed Hyperspectral Images.

Rapid and continuous advancements in remote sensing technology have resulted in finer resolutions and higher acquisition rates of hyperspectral images (HSIs). These developments have triggered a need for new processing techniques brought about by the confined power and constrained hardware resources aboard satellites. This article proposes two novel lossless and near-lossless compression methods, employing our recent seed generation and quadrature-based square rooting algorithms, respectively. The main advantage of the former method lies in its acceptable complexity utilizing simple arithmetic operations, making it suitable for real-time onboard compression. In addition, this near-lossless compressor could be incorporated for hard-to-compress images offering a stabilized reduction at nearly 40% with a maximum relative error of 0.33 and a maximum absolute error of 30. Our results also show that a lossless compression performance, in terms of compression ratio, of up to 2.6 is achieved when testing with hyperspectral images from the Corpus dataset. Further, an improvement in the compression rate over the state-of-the-art k2-raster technique is realized for most of these HSIs by all four variations of our proposed lossless compression method. In particular, a data reduction enhancement of up to 29.89% is realized when comparing their respective geometric mean values.

Self-supervised learning for remote sensing scene classification under the few shot scenario.

Scene classification is a crucial research problem in remote sensing (RS) that has attracted many researchers recently. It has many challenges due to multiple issues, such as: the complexity of remote sensing scenes, the classes overlapping (as a scene may contain objects that belong to foreign classes), and the difficulty of gaining sufficient labeled scenes. Deep learning (DL) solutions and in particular convolutional neural networks (CNN) are now state-of-the-art solution in RS scene classification; however, CNN models need huge amounts of annotated data, which can be costly and time-consuming. On the other hand, it is relatively easy to acquire large amounts of unlabeled images. Recently, Self-Supervised Learning (SSL) is proposed as a method that can learn from unlabeled images, potentially reducing the need for labeling. In this work, we propose a deep SSL method, called RS-FewShotSSL, for RS scene classification under the few shot scenario when we only have a few (less than 20) labeled scenes per class. Under this scenario, typical DL solutions that fine-tune CNN models, pre-trained on the ImageNet dataset, fail dramatically. In the SSL paradigm, a DL model is pre-trained from scratch during the pretext task using the large amounts of unlabeled scenes. Then, during the main or the so-called downstream task, the model is fine-tuned on the labeled scenes. Our proposed RS-FewShotSSL solution is composed of an online network and a target network both using the EfficientNet-B3 CNN model as a feature encoder backbone. During the pretext task, RS-FewShotSSL learns discriminative features from the unlabeled images using cross-view contrastive learning. Different views are generated from each image using geometric transformations and passed to the online and target networks. Then, the whole model is optimized by minimizing the cross-view distance between the online and target networks. To address the problem of limited computation resources available to us, our proposed method uses a novel DL architecture that can be trained using both high-resolution and low-resolution images. During the pretext task, RS-FewShotSSL is trained using low-resolution images, thereby, allowing for larger batch sizes which significantly boosts the performance of the proposed pipeline on the task of RS classification. In the downstream task, the target network is discarded, and the online network is fine-tuned using the few labeled shots or scenes. Here, we use smaller batches of both high-resolution and low-resolution images. This architecture allows RS-FewshotSSL to benefit from both large batch sizes and full image sizes, thereby learning from the large amounts of unlabeled data in an effective way. We tested RS-FewShotSSL on three RS public datasets, and it demonstrated a significant improvement compared to other state-of-the-art methods such as: SimCLR, MoCo, BYOL and IDSSL.

Novel seed generation and quadrature-based square rooting algorithms.

The square root operation is indispensable in a myriad of computational science and engineering applications. Various computational techniques have been devised to approximate its value. In particular, convergence methods employed in this regard are highly affected by the initial approximation of the seed value. Research shows that the provision of an initial approximation with higher accuracy yields fewer additional iterations to calculate the square root. In this article, we propose two novel algorithms. The first one presents a seed generation technique that depends on bit manipulation and whose output is to be used as an initial value in the calculation of square roots. The second one describes a quadrature-based square rooting method that utilizes a rectangle as the plane figure for squaring. We provide error estimation of the former using the vertical parabola equation and employ a suitable lookup table, for the latter, to store needed cosine values. The seed generation approach produces a significant reduction in the number of iterations of up to 84.42% for selected convergence methods. The main advantages of our proposed square rooting algorithm lie in its high accuracy and in its requirement of just a single iteration. Our proposed algorithm also provides for lower computational latency, measured in the number of clock cycles, compared to Newton-Raphson's and Bakhshali's square rooting methods.

A Systematic Review of Hardware-Accelerated Compression of Remotely Sensed Hyperspectral Images.

Hyperspectral imaging is an indispensable technology for many remote sensing applications, yet expensive in terms of computing resources. It requires significant processing power and large storage due to the immense size of hyperspectral data, especially in the aftermath of the recent advancements in sensor technology. Issues pertaining to bandwidth limitation also arise when seeking to transfer such data from airborne satellites to ground stations for postprocessing. This is particularly crucial for small satellite applications where the platform is confined to limited power, weight, and storage capacity. The availability of onboard data compression would help alleviate the impact of these issues while preserving the information contained in the hyperspectral image. We present herein a systematic review of hardware-accelerated compression of hyperspectral images targeting remote sensing applications. We reviewed a total of 101 papers published from 2000 to 2021. We present a comparative performance analysis of the synthesized results with an emphasis on metrics like power requirement, throughput, and compression ratio. Furthermore, we rank the best algorithms based on efficiency and elaborate on the major factors impacting the performance of hardware-accelerated compression. We conclude by highlighting some of the research gaps in the literature and recommend potential areas of future research.

P-RnaPredict--a parallel evolutionary algorithm for RNA folding: effects of pseudorandom number quality.

This paper presents a fully parallel version of RnaPredict, a genetic algorithm (GA) for RNA secondary structure prediction. The research presented here builds on previous work and examines the impact of three different pseudorandom number generators (PRNGs) on the GA's performance. The three generators tested are the C standard library PRNG RAND, a parallelized multiplicative congruential generator (MCG), and a parallelized Mersenne Twister (MT). A fully parallel version of RnaPredict using the Message Passing Interface (MPI) was implemented on a 128-node Beowulf cluster. The PRNG comparison tests were performed with known structures whose sequences are 118, 122, 468, 543, and 556 nucleotides in length. The effects of the PRNGs are investigated and the predicted structures are compared to known structures. Results indicate that P-RnaPredict demonstrated good prediction accuracy, particularly so for shorter sequences.