In inertial confinement fusion (ICF) systems, damage induced by high power lasers is one of the key factors for limiting the further improvement of energy density. Also, there is a close relationship between various defects and damage characteristics of optical components. Usually, the optical components detected in the system have a large-aperture and a variety of surface types. The defects have the characteristics of small scale, low distribution density and wide distribution range, etc. It is a great challenge to achieve multi-faceted detection with high precision and efficiency. In this paper, according to the requirements for micro defects detection on the surface of large-aperture optical components, a multimodal scanning detection method combining the bright-field and dark-field imaging (BDFI) is given. Based on the high sensitivity of dark-field microscopic imaging (DFMI), the ring illumination strategy of collimate light source is studied to locate defects globally. The coaxial illumination method is used to accurately detect the characteristic parameters of defects, which is able to realize the size quantification of multi-morphological defects. At the same time, in order to cope with the detection needs of large-aperture and multi-faceted optical components, the sub-aperture scanning method based on the distribution structure of geodesic sphere is proposed. The multi-aperture scanning path is automatically planned and the spatial distribution of defects is reconstructed with three-dimensional mapping. On this basis, by cooperating with the automatic motion device, automatic focus and high-speed automatic scanning of detection can be realized. The methods proposed in this paper can be accomplished for efficient detection of large-aperture optics with a resolution of 0.5μm for surface defects, which meets the needs of multi-faceted detection. Therefore, it is of great significance to guide the precision machining of optical components and support the process optimization, which has great importance for the final research results of the ICF system.
The classification of scrap steel is the key step in the recycling and utilization of scrap steel. Human detection has been widely used in the classification of scrap steel carriages at present. One major drawback of this approach is the low efficiency of recycling due to the instability of the operator. Therefore, it is necessary to develop a fast and accurate method for automatic classification of scrap steel carriages. This paper proposes an improving method of classifying scrap steel carriages based on deep learning. First, the key frames in the video stream are obtained by the target detection algorithm, then the features of interests are extracted by the feature extraction algorithm, and finally the classification result of the entire carriage is output by the feature fusion algorithm. In the YOLO algorithm for detecting targets, the Darknet network is abandoned and the MobileNet network is used. The spatiotemporal information separation strategy is used when extracting features. The n×1×1 convolution kernel operator is used in the 3D convolutional network of fusion features. In the self-attention network, only the attention mechanism is set for the time dimension. With the analysis of the different sample ratios of the training set and test set, the method proposed in this paper has the characteristics of strong generalization ability, high accuracy, and fast speed which has provided a deeper insight into classification of scrap steel carriages.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.