Automatic traffic-sign detection is a hot topic in computer vision and one of the critical technologies of intelligent transportation. The Transformer structure has recently become a research hotspot due to its excellent performance. We hope to apply this structure to the design of traffic sign detection algorithms. Therefore, we make some improvements to Sparse R-cnn, a neural network model inspired by Transformer. Sparse R-cnn is a novel model, and its core idea is to replace hundreds of thousands of candidate anchors in the RPN network with a small set of proposal boxes. The experiments in our paper proved that the performance of the Sparse R-cnn model is better than other existing general object detection models. Based on the original Sparse R-cnn inspiration, an improved Sparse R-cnn model is proposed. First, a novel backbone for the task of traffic-sign detection is proposed. Multi-scale fusion structure is the essential method of improving the algorithm for small target detection, so improving the multi-scale capability of the backbone is a required method for designing traffic sign detection. So, we made further improvements to the existing backbone ResNest. We enhanced the multi-scale representation ability of the backbone by constructing hierarchical residual-like connections within each single radix block in the original ResNest. We call the improved backbone Res2Nest. The novel backbone proposed by us shows better performance without introducing excessive computational costs to the model. In addition, the attention mechanism is also an effective method to improve the detection of traffic signs, so we set up a branch network for recalibrating the channel feature response adaptively through the Global Average Pooling (GAP) operation and a fully connected layer. It can also be seen as the implementation of the crosschannel self-attention mechanism. After experiments by TT100K dataset, our method would attain a better accuracy and robustness.
In real-time system, the Infrared Focal Plane Array (IRFPA) should be compensated in traditional method of two-point temperature non-uniformity correction (NUC) before system was used every time, which make the system complex. Based on discussion of traditional two-point temperature NUC algorithm, Two-point temperature NUC based on least mean square(LMS)algorithm was proposed in the paper. In the view of the LMS algorithm theory, the correction gain and offset coefficients were iterated one by one during image processing, so that the expected correction result were obtained in little time. At the same time, the correction and coefficients iteration processes were completed in FPGA and DSP respectively, and make the arithmetic structure simple. The simplified structure, low cost and out-door suitable operation-systems are the merits of the system.
To research the attenuation performance of the AlGaN photocathode, three samples with the same structures grown by metalorganic chemical vapor deposition were activated with three different activation methods, which are called Cs-only, Cs-O, and Cs-O-Cs activation, respectively. The spectral responses and attenuated photocurrents of the three AlGaN photocathodes were measured. The results show that the Cs-O activated AlGaN photocathode exhibits the lowest attenuation speed in the first few hours, and the attenuation speed of the Cs-only activated one is fastest. After attenuating for 90 min, the attenuation photocurrent curve of the Cs-O-Cs activated sample is coincident with that of the Cs-O activated one. The main factor affecting the photocurrent attenuation is related to Cs atoms desorbed from the photocathode surface.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.