An Armature Defect Self-Adaptation Quantitative Assessment System Based on Improved YOLO11 and the Segment Anything Model

Processes, Год журнала: 2025, Номер 13(2), С. 532 - 532

Опубликована: Фев. 14, 2025

There is a need to address challenges faced in detecting and segmenting defects micro-vibration motor armatures, which are crucial components used digital devices. Due their complex structure tiny size, quality control during assembly difficult. In this paper, an adaptive segmentation quantization (ASQ) system based on YOLO 11 SAM proposed the issue above. The consists of target detection (TD) unit, shape (SS) quantitative assessment (AS) introduces practical combination YOLO11 for defect segmentation, integrating with novel framework measure severity occurrence. This approach efficient cost-effective, supporting real-time industrial applications by allowing automated, rapid analysis improvement identification. Finally, evaluation standard more than 90% accuracy was achieved. Additionally, hardware developed implement settings. adopts strategy intelligent morphological feature extraction computation, focusing pixel-level assessment. research makes significant step forward automating processes micro-scale components, providing robust solution enhancement manufacturing efficiency product quality.

Язык: Английский

---

SAM-Enhanced Cross-Domain Framework for Semantic Segmentation: Addressing Edge Detection and Minor Class Recognition

Processes, Год журнала: 2025, Номер 13(3), С. 736 - 736

Опубликована: Март 3, 2025

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

Язык: Английский

---

Solarsam: Building-Scale Photovoltaic Potential Assessment Based on Segment Anything Model (Sam) and Remote Sensing for Emerging City

Опубликована: Янв. 1, 2024

Язык: Английский

---

SolarSAM: Building-scale photovoltaic potential assessment based on Segment Anything Model (SAM) and remote sensing for emerging city

Renewable Energy, Год журнала: 2024, Номер 237, С. 121560 - 121560

Опубликована: Окт. 9, 2024

Язык: Английский

---