Wyniki wyszukiwania - Biblioteka Nauki

Ten serwis zostanie wyłączony 2025-02-11.

Nowa wersja platformy, zawierająca wyłącznie zasoby pełnotekstowe, jest już dostępna.
Przejdź na https://bibliotekanauki.pl

Ograniczanie wyników

2 2024

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: YOLOv7

Sortuj według:

Ogranicz wyniki do:

CC-De-YOLO: A Multiscale Object Detection Method for Wafer Surface Defect

100%

Ma J. , Zhang T. , Ma X. , Tian H.

Foundations of Computing and Decision Sciences

2024

tom Vol. 49, No. 3

261--285

Surface defect detection on wafers is crucial for quality control in semiconductor manufacturing. However, the complexity of defect spatial features, including mixed defect types, large scale differences, and overlapping, results in low detection accuracy. In this paper, we propose a CC-De-YOLO model, which is based on the YOLOv7 backbone network. Firstly, the coordinate attention is inserted into the main feature extraction network. Coordinate attention decomposes channel attention into two one-dimensional feature coding processes, which are aggregated along both horizontal and vertical spatial directions to enhance the network’s sensitivity to orientation and position. Then, the nearest neighbor interpolation in the upsampling part is replaced by the CAR-EVC module, which predicts the upsampling kernel from the previous feature map and integrates semantic information into the feature map. Two residual structures are used to capture long-range semantic dependencies and improve feature representation capability. Finally, an efficient decoupled detection head is used to separate classification and regression tasks for better defect classification. To evaluate our model’s performance, we established a wafer surface defect dataset containing six typical defect categories. The experimental results show that the CCDe-YOLO model achieves 91.0% mAP@0.5 and 46.2% mAP@0.5:0.95, with precision of 89.5% and recall of 83.2%. Compared with the original YOLOv7 model and other object detection models, CC-De-YOLO performs better. Therefore, our proposed method meets the accuracy requirements for wafer surface defect detection and has broad application prospects. The dataset containing surface defect data on wafers is currently publicly available on GitHub (https://github.com/ztao3243/Wafer-Datas.git).

The system of headlights operation recognition using the digital twin method

86%

Dawid A. , Buchwald P.

International Journal of Electronics and Telecommunications

2024

tom Vol. 70, No. 1

51--58

Virtual digital representation of a physical object or system, created with precision through computer simulations, data analysis, and various digital technologies can be used as training set for real life situations. The principal aim behind creating a virtual representation is to furnish a dynamic, data-fueled, and digital doppelgänger of the physical asset. This digital counterpart serves multifaceted purposes, including the optimization of performance, the continuous monitoring of its well-being, and the augmentation of informed decision-making processes. Main advantage of employing a digital twin is its capacity to facilitate experimentation and assessment of diverse scenarios and conditions, all without impinging upon the actual physical entity. This capability translates into substantial cost savings and superior outcomes, as it allows for the early identification and mitigation of issues before they escalate into significant problems in the tangible world. Within our research endeavors, we've meticulously constructed a digital twin utilizing the Unity3D software. This digital replica faithfully mimics vehicles, complete with functioning headlamp toggles. Our lighting system employs polygons and normal vectors, strategically harnessed to generate an array of dispersed and reflected light effects. To ensure realism, we've meticulously prepared the scene to emulate authentic road conditions. For validation and testing, we integrated our model with the YOLO (You Only Look Once) neural network. A specifically trained compact YOLO model demonstrated impressive capabilities by accurately discerning the status of real vehicle headlamps. On average, it achieved an impressive recognition probability of 80%, affirming the robustness of our digital twin.