Unsupervised domain adaptation (UDA) enables training a model on labeled source data to perform well in a target domain without supervision, which is especially valuable in vision-based semantic segmentation. However, existing UDA methods often struggle with accurate semantic labeling at object boundaries and recognizing minor categories in the target domain. This paper introduces a novel UDA framework—SamDA—that incorporates the Segment Anything Model (SAM), a large-scale foundational vision model, as the mask generator to enhance edge segmentation performance. The framework comprises three core modules: a cross-domain image mixing module, a self-training module with a teacher–student network, and exponential moving average (EMA). It also includes a finetuning module that leverages SAM-generated masks for pseudo-label matching. Evaluations on the GTA5 and Cityscapes datasets demonstrate that SamDA achieves a mean IoU (mIoU) of 75.2, surpassing state-of-the-art methods such as MIC-DAFormer by 1.0 mIoU and outperforming all ResNet-based approaches by at least 15 mIoU. Moreover, SamDA significantly enhances the segmentation of small objects like bicycles, riders, and fences, with, respective, IoU improvements of 4.5, 5.2, and 3.8 compared to baseline models.

The basic position of microscope in experimental teaching is increasingly prominent, but there is still a big gap between its multidimensional index factors such as operation efficiency, effectiveness, and user satisfaction and the goal of teaching application. Based on the usability design theory, Kinovea video analysis software and ANSYS finite element analysis software were separately used to carry out user observation and product structure testing and analyze the multiple heterogeneous factors affecting the use of microscopes in teaching application scenarios and record usability data; the Fuzzy Analytic Hierarchy Process was utilized to summarize and analyze the usability data and construct a usability evaluation model of microscope. Under the guidance of the usability evaluation model, the microscope was optimized, and its usability was verified by Jack simulation and finite element analysis at last. The results show that: On the layer of operational efficiency, the usability index of intuitive operation and easy to learn of the microscope accounts for the largest weight. On the layer of effectiveness, the index of stable structural connection of the microscope has a larger weight; adjusting the material properties and product structure can effectively improve the stability of the microscope. On the layer of user satisfaction, the index of use comfortably of the microscope accounts for the largest weight, and the comfort of students’ neck and other joints can be improved by optimizing the microscope’s functional layout. This research provides an evaluation index reference and design optimization experience for the multiple heterogeneous factor-driven microscope usability design in teaching applications.