As one of the core applications of computer vision, object detection has become more important in scenarios requiring high accuracy but with limited computational resources such as robotics and autonomous vehicles. Object detection using machine learning running on embedded device such as Raspberry Pi provides the high possibility to detect any custom objects without the recalibration of camera. In this work, we developed a smart and lean object detection model for shipping containers by using the state-of-the-art deep learning TensorFlow model and deployed it to a Raspberry Pi. Using EfficientDet-Lite2, we explored the different cross-validation strategies (Hold-out and K-Fold). The experimental results show that compared with the baseline EfficientDet-Lite2 algorithm, our model improved the mean average precision (mAP) by 44.73% for the Hold-out dataset and 6.26% for K-Fold cross-validation. We achieved Average Precision (AP) of more than 80% and best detection scores of more than 93% for the Hold-out dataset. For the 5-Fold lean dataset, the results show the Average Precision across the three lightweight models are generally high as the models achieved more than 50% average precision, with YOLOv4 Tiny performing better than EfficientDet-Lite2 and Single Shot Detector (SSD) MobileNet V2 Feature Pyramid Network (FPN) 320 as a lightweight model.

No abstract

This paper is an extended paper from the 24th International Conference on Mechatronics Technology, ICMT 2021. The basic mechanical characteristic that gauges the stiffness of a solid material is known as the Young’s modulus. To evaluate the Young’s modulus, destructive material testing is frequently used. This paper describes how to determine a material’s dynamic Young’s modulus using Digital Image Correlation (DIC) in conjunction with numerical back-analysis. Three different materials (brass, aluminum, and steel) were examined for their static and dynamic reactions. A static transverse displacement was first applied at the free end of the beam before it was released and the beam was allowed to vibrate freely. The resulting vibrations at the free end were monitored using the DIC method, following which the natural frequencies of the beam were derived by applying the Fast Fourier Transform (FFT) to the DIC measured time history. The Young’s modulus corresponding to the fundamental natural frequency of the beam was then obtained via modal back-analysis using the finite element program Ansys 2022 R1. In this way, the Young’s modulus of the material may be calculated using a combination of numerical and DIC techniques, thus allowing for the non-contact evaluation of the structural integrity without subjecting the material to destructive testing. Potential applications of this method include bridge and building assessments, and structural health monitoring (SHM).

No abstract

No abstract