With the depletion of high-grade iron ore resources, the efficient utilization of low-grade iron ore has become a critical demand in the steel industry. Due to its uniform particle size and chemical composition, pelletized iron ore significantly enhances both the utilization rate of iron ore and the efficiency of metallurgical processes. This paper presents a deep learning model based on ResUNet, which integrates three-dimensional CT images obtained through industrial computed tomography (ICT) to precisely segment hematite, liquid phase, and porosity. By incorporating residual connections and batch normalization, the model enhances both robustness and segmentation accuracy, achieving F1 scores of 98.37%, 95.10%, and 83.87% for the hematite, pores, and liquid phase, respectively, on the test set. Through 3D reconstruction and quantitative analysis, the volume fractions and fractal dimensions of each component were computed, revealing the impact of the spatial distribution of different components on the physical properties of the pellets. Systematic evaluation of model robustness demonstrated varying sensitivity to different CT artifacts, with the strongest resistance to beam hardening and highest sensitivity to Gaussian noise. Multi-scale resolution analysis revealed that segmentation quality and fractal dimension estimates exhibit phase-dependent responses to resolution changes, with the liquid phase being the most sensitive. Despite these dependencies, the relative complexity relationships among phases remained consistent across resolutions, supporting the reliability of our qualitative conclusions. The study demonstrates that the deep learning-based image segmentation method effectively captures microstructural details, reduces human error, and enhances automation, providing a scientific foundation for optimizing pellet quality and improving metallurgical performance. It holds considerable potential for industrial applications.

In order to help rural agriculture achieve maximum economic benefits and controllable risks, this paper focuses on the optimization of crop planting strategies in a certain village from 2024 to 2030 and constructs a series of models. Firstly, simulated annealing and mixed-integer linear programming algorithms are used to optimize different scenarios of crop overproduction under stable assumptions, and the optimal planting plan is determined. It is found that the strategy of selling overproduced crops at a low price is more beneficial. Then, Monte Carlo simulation and sequential least-squares programming are introduced. By considering the uncertainty of multiple factors and improving the objective function, the planting plan is further optimized. The research results provide a scientific basis for rural agricultural planting decisions, demonstrate the effectiveness of models and algorithms in agricultural planning, and are of great significance for promoting the sustainable development of rural economy.