Object detection in aerial images is a growing area of research, with maritime object detection being a particularly important task for reliable surveillance, monitoring, and active rescuing. Notwithstanding astonishing advances of computer visiontechnologies, detecting ships and floating matters in these images are challenging due to factors such as object distance. What makes it worse is pervasive sea surface effects such as sunlight reflection, wind, and waves. Hyperspectral image (HSI) sensors, providing more than 100 channels in wavelengths of visible and near-infrared, can extract intrinsic information of materials from a few pixels of HSIs.The advent of HSI sensors motivates us to leverage HSIs to circumvent false positives due to the sea surface effects.Unfortunately, there are few public HSI datasets due to the high cost and labor involved in collecting them, hindering object detection research based on HSIs. We have collected and annotated a new dataset called ``Multi-Modal Ship and flOating matter Detection in Aerial Images (M$^{2}$SODAI),'', which includes synchronized image pairs of RGB and HSI data, along with bounding box labels for nearly 6,000 instances per category. We also propose a new multi-modal extension of the feature pyramid network called DoubleFPN.Extensive experiments on our benchmark demonstrate that fusion of RGB and HSI data can enhance mAP, especially in the presence of the sea surface effects.

BACKGROUND: Protein secondary structures that link simple 1D sequences to complex 3D structures can be used as good features for describing the local properties of protein, but also can serve as key features for predicting the complex 3D structures of protein. Thus, it is very important to accurately predict the secondary structure of the protein, which contains a local structural property assigned by the pattern of hydrogen bonds formed between amino acids. In this study, we accurately predict protein secondary structure by capturing the local patterns of protein. For this objective, we present a novel prediction model, AttSec, based on transformer architecture. In particular, AttSec extracts self-attention maps corresponding to pairwise features between amino acid embeddings and passes them through 2D convolution blocks to capture local patterns. In addition, instead of using additional evolutionary information, it uses protein embedding as an input, which is generated by a language model. RESULTS: For the ProteinNet DSSP8 dataset, our model showed 11.8% better performance on the entire evaluation datasets compared with other no-evolutionary-information-based models. For the NetSurfP-2.0 DSSP8 dataset, it showed 1.2% better performance on average. There was an average performance improvement of 9.0% for the ProteinNet DSSP3 dataset and an average of 0.7% for the NetSurfP-2.0 DSSP3 dataset. CONCLUSION: We accurately predict protein secondary structure by capturing the local patterns of protein. For this objective, we present a novel prediction model, AttSec, based on transformer architecture. Although there was no dramatic accuracy improvement compared with other models, the improvement on DSSP8 was greater than that on DSSP3. This result implies that using our proposed pairwise feature could have a remarkable effect for several challenging tasks that require finely subdivided classification. Github package URL is https://github.com/youjin-DDAI/AttSec .

In interventional studies, biomarkers such as metabolites, are usually measured across serial time points. And when the interest lies in comparing expression levels between different experimental conditions, summary measures such as area under curve (AUC), have been widely used. Although the summary measure based approaches have been successful in identifying novel biomarkers, they do not reveal anything about time-dependent changing patterns of biomarkers which can demonstrate the reactivity of biomarkers to various physiological conditions. To account for such patterns, all measurements across time points need to be used, and clustering analysis with the measurements can group together biomarkers having similar changing patterns. Some such popularly used clustering methods include hierarchical- and K-means clustering. While these may provide some well-clustered results, their patterns are quite dependent on input data sets, making it difficult to obtain consistent patterns across different interventional studies. In addition, it is problematic for these methods to discriminate biomarkers with weakly active patterns that need to be grouped as static, compared to those having strongly active patterns, when their patterns are highly similar. To address these issues, we propose a new clustering method for improving identification of changing patterns. Our approach is based on a two-stage process: the first is elimination of stable markers using Euclidean distances, while the second stage assigns the remaining biomarkers to predefined patterns using 1-correlation distance measure. By simulation studies, we showed that our proposed method had superior classification performances, compared to other unsupervised clustering methods. We expect that this approach can complement the existing summary measure based approaches.