The maize field environment is complex. Weeds and maize have similar colors and may overlap, and lighting and weather conditions vary. Thus, many methods for the automated differentiation of maize and weeds achieve poor segmentation or cannot be used in real time. In this paper, a weed recognition model based on improved Swin-Unet is proposed. The model first performs semantic segmentation of maize seedlings and uses the resulting mask to identify weeds. U-Net acts as the semantic segmentation framework, and a Swin transformer module is introduced to improve performance. DropBlock regularization, which randomly hides some blocks in crop feature maps, is applied to enhance the generalization ability of the model. Finally, weed areas are identified and segmented with the aid of an improved morphological processing algorithm. The DeepLabv3+, PSANet, Mask R-CNN, original Swin-Unet, and proposed models are trained on a dataset of maize seedling images. The proposed Swin-Unet model outperforms the others, achieving a mean intersection over union of 92.75%, mean pixel accuracy of 95.57%, and inference speed of 15.1 FPS. Our model could be used for accurate, real-time segmentation of crops and weeds and as a reference for the development of intelligent agricultural equipment.

No abstract

Background: Electromagnetic wave absorbers that has near perfect performance have been studied since 2008, various one dimensional and two dimensional structures have been optimized. High absorption at certain frequency can be easily realized. However, high absorption that covers broad band is relatively hard to realized. Method: In this paper, a nickel nanograting is designed by using rigorous coupled wave analysis theory. The effect of grating height and grating width on absorption performance is analyzed physically by using equivalent medium theory and Fabry-Pérot resonate theory. Results: Because of the intrinsic absorption character and the influence of F-P resonance, TM absorption shows periodically increase with the increasing of metal thickness. Absorption of the TE polarization slightly increases with the increasing of grating height. Peak absorption of TM polarization light can be gotten when DC of 0.52 is chosen, however, absorption performance of TE polarization light degrades seriously. In this paper, DC of 0.19 is chosen. Conclusion: A one dimensional nickel grating based broadband absorber is analyzed in this paper. The proposed absorber has good absorption performance over the visible band, and shows no clear degradation over a wide incident angle (-40° to 40°).

Using equivalent medium theory, surface plasma resonance theory and Fabry-Pérot resonance theory, transmission mechanism of single-layer nanowire polarizer is systematically analyzed. Concept of skin depth is used to explain the effect of grating material on polarization transmission. Two transmission minimum points in the curve of grating period versus transverse magnetic transmission are analyzed quantitatively. Effects of grating duty cycle, height and type on polarization transmission of single-layer nanowire polarizer are also analyzed. The requirements of designing single-layer sub-wavelength metal gratings with good polarization properties are given. The analysis of the transmission mechanism of sub-wavelength metal gratings will be helpful for the design and fabrication of the gratings with good polarization properties.

No abstract