Surface Defect Detection of Vehicle Light Guide Plates Based on an Improved RetinaNet

Aiming at the vehicle navigation light guide plate (LGP) image characteristics, such as complex and gradient texures, uneven brightness, and small defects, this paper proposes a visual inspection method for LGP defects based on an improved RetinaNet. First, we use ResNeXt50 with higher accuracy under the same parameters as the backbone network, and propose the lightweight module Ghost_module to replace the 1×1 convolution in the lower half of the ResNeXt_block. This can reduce the resource parameters and consumption, and speed up training and inference. Second, we propose and use an improved feature pyramid network (IFPN) module to improve the feature fusion network in RetinaNet. It can more effectively fuse the shallow semantic information and high-level semantic information in the backbone feature extraction network, and further improve the detection ability of small target defects. Finally, the defect detection dataset constructed based on the vehicle LGP images collected at a industrial site, and experiments are performed on the vehicle LGP dataset and Aluminum Profile Defect Identification dataset (Aluminum Profile DID). The experimental results show that the proposed method is both efficient and effective. It achieves a better average detection rate of 98.6% on the vehicle LGP dataset. The accuracy and real-time performance can meet the requirements of industrial detection.

Optimized Micro-Pattern Design and Fabrication of a Light Guide Plate Using Micro-Injection Molding

This study examined the uniformity of illuminance field distributions of light guide plates (LGPs). First, the authors designed microstructural patterns on the surface of an LGP. Then, a mold of the LGP with the optimal microstructural design was fabricated by a photolithography method. Micro-injection molding (μIM) was used to manufacture the molded LGPs. μIM technology can simultaneously manufacture large-sized wedge-shaped LGPs and micro-scale microstructures. Finally, illuminance values of the field distributions of the LGPs with various microstructures were obtained through optical field measurements. This study compared the illuminance field distributions of LGPs with various designs and structures, which included LGPs without and those with microstructure on the primary design and the optimal design. The average illuminance of the LGP with microstructures and the optimal design was roughly 196.1 cd/m2. Its average illuminance was 1.3 times that of the LGP without microstructures. This study also discusses illuminance field distributions of LGPs with microstructures that were influenced by various μIM process parameters. The mold temperature was found to be the most important processing parameter affecting the illuminance field distribution of molded LGPs fabricated by μIM. The molded LGP with microstructures and the optimal design had better uniformity than that with microstructures and the primary design and that without microstructures. The uniformity of the LGP with microstructures and the optimal design was roughly 86.4%. Its uniformity was nearly 1.65 times that of the LGP without microstructures. The optimized design and fabrication of LGPs with microstructure exhibited good uniformity of illuminance field distributions.

The development of an optical system for lighting various rooms with sunlight

The necessity of using light in the room, the spectral composition of which corresponds to the spectrum of the sun for a comfortable state of a person, is justified. A passive system that does not consume energy is proposed. An energy-efficient natural light system has been developed, creating a comfortable environment for the eyes, well-being and mood of a person. An experimental study of the system operation was carried out. The dependence of the light intensity on the angle of incidence in the light guide is obtained.

Dynamic Measurements of Tank Wall Deformations as a New Method for Assessing its Technical Condition

The paper proposes a new approach to assessing the residual resource of the wall of a vertical steel tank by the criterion of low-cycle loading, taking into account its actual deformations. The method is based on the principle of measuring the difference in the power of the luminous flux at the ends of the light tube (light guide) when it is bent. An experiment was carried out using an optical pair (LED and photodiode) and an optical fiber. For this purpose, templates with known radii of curvature were prepared, a prototype of an Arduino-based device was developed and an optical fiber was manufactured, which was subsequently fixed on the wall of a two-hundred-liter metal barrel. As a result of the experiments, a formula was obtained for determining the curvature of the wall depending on the readings of the sensor. The sensitivity of the proposed method is estimated. An algorithm for determining the number of cycles before the formation of fatigue cracks is proposed, taking into account the actual values of the wall deflection.