Rapid super resolution for infrared imagery.

Infrared (IR) imagery is used in agriculture for irrigation monitoring and early detection of disease in plants. The common IR cameras in this field typically have low resolution. This work offers a method to obtain the super-resolution of IR images from low-power devices to enhance plant traits. The method is based on deep learning (DL). Most calculations are done in the low-resolution domain. The results of each layer are aggregated together to allow a better flow of information through the network. This work shows that good results can be achieved using depthwise separable convolution with roughly 300K multiply-accumulate computations (MACs), while state-of-the-art convolutional neural network-based super-resolution algorithms are performed with around 1500K MACs. MTF analysis of the proposed method shows a real ×4 improvement in the spatial resolution of the system, out-preforming the diffraction limit. The method is demonstrated on real agricultural images.

[1]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[2]  Michael Elad,et al.  Fast and robust multiframe super resolution , 2004, IEEE Transactions on Image Processing.

[3]  Victor Alchanatis,et al.  Image fusion of visible and thermal images for fruit detection. , 2009 .

[4]  Thomas S. Huang,et al.  Image Super-Resolution Via Sparse Representation , 2010, IEEE Transactions on Image Processing.

[5]  C. R. Jalwania,et al.  Uncooled Infrared Microbolometer Arrays and their Characterisation Techniques (Review Paper) , 2009 .

[6]  Yanlong Cao,et al.  Cascaded Deep Networks With Multiple Receptive Fields for Infrared Image Super-Resolution , 2019, IEEE Transactions on Circuits and Systems for Video Technology.

[7]  Kwang In Kim,et al.  Single-Image Super-Resolution Using Sparse Regression and Natural Image Prior , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[8]  Tianyi Wang,et al.  Terahertz image super-resolution based on a deep convolutional neural network. , 2019, Applied optics.

[9]  Xiaoou Tang,et al.  Image Super-Resolution Using Deep Convolutional Networks , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[10]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[11]  Tomer Michaeli,et al.  Deep-STORM: super-resolution single-molecule microscopy by deep learning , 2018, 1801.09631.

[12]  Peng Fei,et al.  High-throughput, high-resolution deep learning microscopy based on registration-free generative adversarial network. , 2019, Biomedical optics express.

[13]  M. Tester,et al.  High-throughput shoot imaging to study drought responses. , 2010, Journal of experimental botany.

[14]  Ibrahim Abdulhalim,et al.  Tunable extended depth of field using a liquid crystal annular spatial filter. , 2014, Optics letters.

[15]  Liesbet Lagae,et al.  Pixel super-resolution for lens-free holographic microscopy using deep learning neural networks. , 2019, Optics express.

[16]  Lopamudra Mukherjee,et al.  Convolutional neural networks for whole slide image superresolution. , 2018, Biomedical optics express.

[17]  Yan Zhu,et al.  Photon-limited face image super-resolution based on deep learning. , 2018, Optics express.

[18]  Jong Beom Ra,et al.  Resolution Improvement of Infrared Images Using Visible Image Information , 2011, IEEE Signal Processing Letters.

[19]  Y. Cohen,et al.  Use of thermal and visible imagery for estimating crop water status of irrigated grapevine. , 2006, Journal of experimental botany.