No abstract

Label-free biosensor operating within the terahertz (THz) spectra have helped to unlock a myriad of potential terahertz applications, ranging from bio-material detection to point-of-care (PoC) diagnostics. However, the THz wave diffraction limit and the lack of emitter-integrated THz biosensor are hindering the proliferation of high resolution near-field label-free THz biosensing.Here, a monolithic THz emission biosensor is achieved for the first time by integrating asymmetric double-split ring resonator metamaterials with a ferromagnetic heterojunction spintronic THz emitter. This device exhibits an electromagnetically induced transparency window with resonance frequency of 1.02 THz and a spintronic THz radiation source with a bandwidth of 900 GHz, which are integrated on a fused silica substrate monolithically for the first time. It was observed that the Page 2 of 23 ACS Paragon Plus Environment ACS Applied Materials & Interfaces 3resonance frequency experienced a redshift behavior along with increasing the concentration of Hela cells and Pseudomonas due to the strong interaction between the spintronic THz radiation and the biological samples on the metamaterials. The spatial frequency redshift resolution is ~ 0.01 THz with a pseudomonas concentration increase from ~ 0.5×10 4 /mL to ~ 1×10 4 /mL. The monolithic THz biosensor is also sensitive to the sample concentration distribution with 15.68 sensitivity under spatial resolution of 500 μm, which is determined by the infrared pump light diffraction limit. This THz emission biosensor shows great potential for high resolution near-field biosensing applications of trace biological samples.

The existing registration algorithms suffer from low precision and slow speed when registering a large amount of point cloud data. In this paper, we propose a point cloud registration algorithm based on feature extraction and matching; the algorithm helps alleviate problems of precision and speed. In the rough registration stage, the algorithm extracts feature points based on the judgment of retention points and bumps, which improves the speed of feature point extraction. In the registration process, FPFH features and Hausdorff distance are used to search for corresponding point pairs, and the RANSAC algorithm is used to eliminate incorrect point pairs, thereby improving the accuracy of the corresponding relationship. In the precise registration phase, the algorithm uses an improved normal distribution transformation (INDT) algorithm. Experimental results show that given a large amount of point cloud data, this algorithm has advantages in both time and precision.

No abstract