Abstract Satellite image classification has many important applications that play a crucial role in the areas of urban planning, agriculture, as well as environmental monitoring. Nevertheless, the high accuracy and interpretability of deep learning models with such complex datasets is still a big challenge. To solve this, a new hybrid deep-learning architecture, MRCL-ELM is proposed to improve the performance of satellite image classification. The model uses EfficientNetB0 as its building block in terms of ability to learn rich features in an efficient manner with optimization of the network depth and size to minimize memory and processing requirements. It combines the Multi-Residual Convolutional (MRC) networks to learn spatial features robustly with the aid of multiple residual paths, in each block, in MRC, to enhance the learning of features and gradient flow. It uses a Long Short-Term Memory (LSTM) time series modeling layer, and an Extreme Learning Machine (ELM) to quickly and non-iteratively classify data and is therefore lightweight, accurate, and more scalable than other common deep learning architectures. To enhance the interpretability of the proposed model, Local Interpretable Model-agnostic Explanations (LIME) explains individual predictions by testing small variations in the input, whereas SHapley Additive exPlanations (SHAP) provides feature importance scores throughout the model, along with improving model interpretability and trust. The proposed model provides the highest possible results, with 98.33% accuracy on the EuroSAT dataset and 98.10% accuracy on the UC Merced Land Use dataset, being higher than the use of existing Convolutional Neural Networks (CNN) and transformer-based techniques. The training using fixed random seeds and 5-fold cross-validation is used to ensure robustness. Lastly, MRCL-ELM was implemented as a real-time web-based application and tested with real-life Google Maps imagery, and thus needs real-time, precise, and interpretable satellite image classification for the end-users.

Abstract Accurate precipitation forecasting is crucial for evaluating various hydrological processes. This research explores the application of deep learning models for rainfall prediction in the northern region of Bangladesh, focusing on the comparative performance of six models: CNN (Convolutional Neural Network), LSTM (Long Short-Term Memory), XGB (Extreme Gradient Boosting), Ensemble Model, Transformer-XGB, and CNN-XGB. Two distinct datasets were utilized to assess the effectiveness of these models. Among them, the CNN-XGB hybrid model consistently demonstrated superior performance across all evaluation metrics, establishing it as the most reliable predictor in this study. Rajshahi district’s satellite dataset showed an RMSE (Root Mean Squared Error) of 0.65 mm/day, MAE (Mean Absolute Error) of 0.28 mm/day, and R 2 of 0.99. In the ground dataset, Rajshahi district beat other models with an RMSE of 16.28 mm/month, MAE of 7.85 mm/month, and R 2 of 0.98. These findings demonstrate the model’s efficacy across several data sources. To enhance the interpretability of the proposed CNN-XGB model, we deployed the SHAP (Shapley Additive exPlanations) explainer, providing insights into the model’s decision-making process. This research highlights the potential of hybrid models in enhancing rainfall prediction accuracy while providing transparency through explainable AI techniques. Beyond hydrology, the predicted rainfall patterns provide essential inputs for urban planners to optimize land-use zoning in flood-prone areas, and guide resilient infrastructure development. Code Available: https://github.com/Shafi3397/Rainfall-Prediction-using-CNN-XGBoost

Since its major release in 2006, the Unified Extensible Firmware Interface (UEFI) has become the industry standard for interfacing a computer's hardware and operating system, replacing BIOS. UEFI has higher privileged security access to system resources than any other software component, including the system kernel. Hence, identifying and characterizing vulnerabilities in UEFI is extremely important for computer security. However, automated detection and characterization of UEFI vulnerabilities is a challenging problem. Static vulnerability analysis techniques are scalable but lack precision (reporting many false positives), whereas symbolic analysis techniques are precise but are hampered by scalability issues due to path explosion and the cost of constraint solving. In this paper, we introduce a technique called STatic Analysis guided Symbolic Execution (STASE), which integrates both analysis approaches to leverage their strengths and minimize their weaknesses. We begin with a rule-based static vulnerability analysis on LLVM bitcode to identify potential vulnerability targets for symbolic execution. We then focus symbolic execution on each target to achieve precise vulnerability detection and signature generation. STASE relies on the manual specification of reusable vulnerability rules and attacker-controlled inputs. However, it automates the generation of harnesses that guide the symbolic execution process, addressing the usability and scalability of symbolic execution, which typically requires manual harness generation to reduce the state space. We implemented and applied STASE to the implementations of UEFI code base. STASE detects and generates vulnerability signatures for 5 out of 9 recently reported PixieFail vulnerabilities and 13 new vulnerabilities in Tianocore's EDKII codebase.

Starting with a random initial seed, fuzzers search for inputs that trigger bugs or vulnerabilities. However, fuzzers often fail to generate inputs for program paths guarded by restrictive branch conditions. In this paper, we show that by first identifying rare-paths in programs (i.e., program paths with path constraints that are unlikely to be satisfied by random input generation), and then, generating inputs/seeds that trigger rare-paths, one can improve the coverage of fuzzing tools. In particular, we present techniques 1) that identify rare paths using quantitative symbolic analysis, and 2) generate inputs that can explore these rare paths using path-guided concolic execution. We provide these inputs as initial seed sets to three state of the art fuzzers. Our experimental evaluation on a set of programs shows that the fuzzers achieve better coverage with the rare-path based seed set compared to a random initial seed.