Ultra‐broadband THz photonics covering the 0.3–20 THz range provides a very attractive foundation for a wide range of basic research and industrial applications. However, the lack of ultra‐broadband THz devices has yet to be overcome. In this work, high‐density organic electro‐optic crystals are newly developed for efficient THz wave generation in a very broad THz spectral range and are successfully used for a broadband THz time‐domain spectroscopy. The new organic THz generator crystals, namely the OHP‐TFS crystals, have very low void volume, high density, and are shown to cover the ultra‐broadband THz spectrum up to about 15 THz, which cannot be easily accessed with the more widely used inorganic‐based THz generators. In addition to the very favorable broadband properties, the generated THz electric‐field amplitude at the pump wavelength of 1560 nm is about 40 times higher than that generated by a commercial inorganic THz generator (ZnTe crystal). By using the newly developed OHP‐TFS as generation crystal in a compact table‐top all‐organic THz time‐domain spectrometer based on a low‐cost telecom fiber laser, the optical characteristics of a model material are successfully determined in the broad 1.5–12.5 THz range with high accuracy.
Weakly-supervised semantic segmentation (WSSS) is introduced to narrow the gap for semantic segmentation performance from pixel-level supervision to image-level supervision. Most advanced approaches are based on class activation maps (CAMs) to generate pseudo-labels to train the segmentation network. The main limitation of WSSS is that the process of generating pseudo-labels from CAMs that use an image classifier is mainly focused on the most discriminative parts of the objects. To address this issue, we propose Puzzle-CAM, a process that minimizes differences between the features from separate patches and the whole image. Our method consists of a puzzle module and two regularization terms to discover the most integrated region in an object. Puzzle-CAM can activate the overall region of an object using image-level supervision without requiring extra parameters. In experiments, Puzzle-CAM outperformed previous state-of-the-art methods using the same labels for supervision on the PASCAL VOC 2012 dataset. Code associated with our experiments is available at https://github.com/ OFRIN/PuzzleCAM.
Efficient broadband organic terahertz (THz) generators using X‐shaped alignment of the nonlinear optical chromophores, as an alternative to the parallel alignment of chromophores in benchmark organic crystals, are reported. All the developed six organic benzothiazolium crystals exhibit an isomorphic X‐shaped alignment of chromophores, resulting in an unprecedentedly large off‐diagonal optical nonlinearity (>100 × 10−30 esu), which presents one of the largest off‐diagonal optical nonlinearity of organic crystals. The benzothiazolium crystals exhibit efficient broadband THz wave generation employing the off‐diagonal optical nonlinearity, in contrast to the present state‐of‐the‐art organic THz generators that mostly utilize diagonal optical nonlinearity. For using off‐diagonal and diagonal optical nonlinearities, the polarization of the optical pump is perpendicular and parallel, respectively, to the polar axis of crystals. In addition to a large THz wave generation efficiency with one order of magnitude higher peak‐to‐peak THz electric field than that generated in a 1.0‐mm‐thick inorganic benchmark ZnTe crystal, the benzothiazolium crystals generate broadband THz spectra with an upper cut‐off frequency of near 8 THz, and the absence of strong absorption dimples in the range of 0.5−4 THz. Therefore, the X‐shaped alignment of chromophores presents an interesting potential alternative for efficient broadband organic THz generators.
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.