Two Individual Super-Bound State Modes within Band Gap with Ultra-High Q Factor for Potential Sensing Applications in the Terahertz Wave Band.

Bound states in the continuum (BICs) garnered significant research interest in the field of sensors due to their exceptionally high-quality factors. However, the wide-band continuum in BICs are noise to the bound states, and it is difficult to control and filter. Therefore, we constructed a top-bottom symmetric cavity containing three high permittivity rectangular columns. The cavity supports a symmetry-protected (SP) superbound state (SBS) mode and an accidental (AC) SBS mode within the bandgap. With a period size of 5 × 15, the bandgap effectively filters out the continuum, allowing only the bound states to exist. This configuration enabled us to achieve a high signal-to-noise ratio and a wide free-spectral-range. The AC SBS and the SP SBS can be converted into quasi-SBS by adjusting different parameters. Consequently, the cavity can function as a single-band sensor or a dual-band sensor. The achieved bulk sensitivity was 38 µm/RIU in terahertz wave band, and a record-high FOM reached 2.8 × 108 RIU-1. The effect of fabrication error on the performance for sensor application was also discussed, showing that the application was feasible. Moreover, for experimental realization, a 3D schematic was presented. These achievements pave the way for compact, high-sensitivity biosensing, multi-wavelength sensing, and other promising applications.

Multiple Bound State Soliton Pulses in the All Polarization Maintaining Fiber Laser.

The bound state soliton pulse, a novel mode-locked output state of fiber lasers, has been studied extensively to gain a better understanding of soliton interactions and to explain the mechanism behind the generation of mode-locked pulses. In this particular research, we utilized a self-made saturable absorber (SA) consisting of single-walled carbon nanotubes (SWCNT) in a fully polarization maintaining (PM) erbium-doped fiber optical path. Through this setup, we observed various bound state pulse phenomena, including the double bound state with different phase differences, the bound state formed by two double pulse bound states, the multi-pulse bound state, etc. The abundant bound soliton pulse states demonstrated the excellent nonlinear absorption characteristics of the SA as well as the excellent optical properties of the all-PM fiber laser. It contributed to exploring the relationship between sub pulses and mode-locked pulses in the future. Additionally, due to the strong interaction between bound state solitons and the inherent stability of the PM optical path, there was potential for utilizing this setup as a seed source to enhance the stability of high-power fiber lasers.

Superbound state in photonic bandgap and its application to generate complete tunable SBS-EIT, SBS-EIR and SBS-Fano.

Bound states in continua (BICs) have high-quality factors that may approach infinity. However, the wide-band continua in BICs are noise to the bound states, limiting their applications. Therefore, this study designed fully controlled superbound state (SBS) modes in the bandgap with ultra-high-quality factors approaching infinity. The operating mechanism of the SBS is based on the interference of the fields of two phase-opposite dipole sources. Quasi-SBSs can be obtained by breaking the cavity symmetry. The SBSs can also be used to produce high-Q Fano resonance and electromagnetically-induced-reflection-like modes. The line shapes and the quality factor values of these modes could be controlled separately. Our findings provide useful guidelines for the design and manufacture of compact and high-performance sensors, nonlinear effects, and optical switches.

Tunable bilayer dielectric metasurface via stacking magnetic mirrors.

Functional tunability, environmental adaptability, and easy fabrication are highly desired properties in metasurfaces. Here we provide a tunable bilayer metasurface composed of two stacked identical dielectric magnetic mirrors. The magnetic mirrors are excited by the interaction between the interference of multipoles of each cylinder and the lattice resonance of the periodic array, which exhibits nonlocal electric field enhancement near the interface and high reflection. We achieve the reversible conversion between high reflection and high transmission by manipulating the interlayer coupling near the interface between the two magnetic mirrors. Controlling the interlayer spacing leads to the controllable interlayer coupling and scattering of meta-atom. The magnetic mirror effect boosts the interlayer coupling when the interlayer spacing is small. Furthermore, the high transmission of the bilayer metasurface has good robustness due to the meta-atom with interlayer coupling can maintain scattering suppression against positional perturbation. This work provides a straightforward method to design tunable metasurface and sheds new light on high-performance optical switches applied in communication and sensing.

Geodesic Basis Function Neural Network.

In the learning of existing radial basis function neural networks-based methods, it is difficult to propagate errors back. This leads to an inconsistency between the learning and recognition task. This article proposes a geodesic basis function neural network with subclass extension learning (GBFNN-ScE). The geodesic basis function (GBF), which is defined here for the first time, uses the geodetic distance in the manifold as a measure to obtain the response of the sample with respect to the local center. To learn network parameters by back-propagating errors for the purpose of correct classification, a specific GBF based on a pruned gamma encoding cosine function is constructed. This function has a concise and explicit expression on the hyperspherical manifold, which is conducive to the realization of error back propagation. In the preprocessing layer, a sample unitization method with no loss of information, nonnegative unit hyperspherical crown (NUHC) mapping, is proposed. The sample can be mapped to the support set of the GBF. To alleviate the problem that one-hot encoding is not effective enough in the differential expression of data labels within a class, a subclass extension (ScE) learning strategy is proposed. The ScE learning strategy can help the learned network be more robust. For the working of GBFNN-ScE, the original sample is projected onto the support set of GBF through the NUHC mapping. Then the mapped samples are sent to the nonlinear computation units composed of GBFs in the hidden layer. Finally, the response obtained in the hidden layer is weighted by the learned weight to obtain the network output. This article theoretically proves that the separability of the data with ScE learning is stronger. The experimental results show that the proposed GBFNN-ScE has a better performance in recognition tasks than existing methods. The ablation experiments show that the ideas of the GBFNN-ScE contribute to the algorithm performance.

Elongated-Hexagonal Photonic Crystal for Buffering, Sensing, and Modulation.

A paradigm for high buffering performance with an essential fulfillment for sensing and modulation was set forth. Through substituting the fundamental two rows of air holes in an elongated hexagonal photonic crystal (E-PhC) by one row of the triangular gaps, the EPCW is molded to form an irregular waveguide. By properly adjusting the triangle dimension solitary, we fulfilled the lowest favorable value of the physical-size of each stored bit by about µ5.5510 µm. Besides, the EPCW is highly sensitive to refractive index (RI) perturbation attributed to the medium through infiltrating the triangular gaps inside the EPCW by microfluid with high RI sensitivity of about 379.87 nm/RIU. Furthermore, dynamic modulation can be achieved by applying external voltage and high electro-optical (EO) sensitivity is obtained of about 748.407 nm/RIU. The higher sensitivity is attributable to strong optical confinement in the waveguide region and enhanced light-matter interaction in the region of the microfluid triangular gaps inside the EPCW and conventional gaps (air holes). The EPCW structure enhances the interaction between the light and the sensing medium.

10 µJ noise-like pulse generated from all fiberized Tm-doped fiber oscillator and amplifier.

Herein, we presented a high energy noise-like (NL) pulse Tm-doped fiber laser (TDFL) system. Relying on the nonlinear amplifying loop mirror (NALM), stable noise-like pulses with coherence spike width of ∼317 fs and envelope width of ∼4.2 ns were obtained from an all polarization-maintaining fiberized oscillator at central wavelength of ∼1946.4 nm with 3 dB bandwidth of ∼24.9 nm. After the amplification in an all-fiberized TDF amplifier system, the maximum average output power of ∼32.8 W and pulse energy of ∼10.1 µJ were obtained, which represents the highest pulse energy of NL pulse at ∼2 µm, to the best of our knowledge. We believe that the high energy NL pulse source has the potential application in mid-infrared supercontinuum generation.

Deep Non-Negative Matrix Factorization Architecture Based on Underlying Basis Images Learning.

The non-negative matrix factorization (NMF) algorithm represents the original image as a linear combination of a set of basis images. This image representation method is in line with the idea of "parts constitute a whole" in human thinking. The existing deep NMF performs deep factorization on the coefficient matrix. In these methods, the basis images used to represent the original image is essentially obtained by factorizing the original images once. To extract features reflecting the deep localization characteristics of images, a novel deep NMF architecture based on underlying basis images learning is proposed for the first time. The architecture learns the underlying basis images by deep factorization on the basis images matrix. The deep factorization architecture proposed in this paper has strong interpretability. To implement this architecture, this paper proposes a deep non-negative basis matrix factorization algorithm to obtain the underlying basis images. Then, the objective function is established with an added regularization term, which directly constrains the basis images matrix to obtain the basis images with good local characteristics, and a regularized deep non-negative basis matrix factorization algorithm is proposed. The regularized deep nonlinear non-negative basis matrix factorization algorithm is also proposed to handle pattern recognition tasks with complex data. This paper also theoretically proves the convergence of the algorithm. Finally, the experimental results show that the deep NMF architecture based on the underlying basis images learning proposed in this paper can obtain better recognition performance than the other state-of-the-art methods.

SnSe Nanosheets: From Facile Synthesis to Applications in Broadband Photodetections.

In recent years, using two-dimensional (2D) materials to realize broadband photodetection has become a promising area in optoelectronic devices. Here, we successfully synthesized SnSe nanosheets (NSs) by a facile tip ultra-sonication method in water-ethanol solvent which was eco-friendly. The carrier dynamics of SnSe NSs was systematically investigated via a femtosecond transient absorption spectroscopy in the visible wavelength regime and three decay components were clarified with delay time of τ1 = 0.77 ps, τ2 = 8.3 ps, and τ3 = 316.5 ps, respectively, indicating their potential applications in ultrafast optics and optoelectronics. As a proof-of-concept, the photodetectors, which integrated SnSe NSs with monolayer graphene, show high photoresponsivities and excellent response speeds for different incident lasers. The maximum photo-responsivities for 405, 532, and 785 nm were 1.75 × 104 A/W, 4.63 × 103 A/W, and 1.52 × 103 A/W, respectively. The photoresponse times were ~22.6 ms, 11.6 ms, and 9.7 ms. This behavior was due to the broadband light response of SnSe NSs and fast transportation of photocarriers between the monolayer graphene and SnSe NSs.

High-power mode-locked thulium-doped fiber laser with tungsten ditelluride as saturable absorber.

A passively mode-locked thulium-doped fiber laser using a tungsten ditelluride saturable absorber (${{\rm WTe}_2}\mbox{-}{\rm SA}$WTe2-SA) is demonstrated. High-power mode-locked pulses with an average output power of 108.1 mW were achieved by incorporating the ${{\rm WTe}_2}\mbox{-}{\rm SA}$WTe2-SA into a thulium-doped fiber oscillator. To the best of our knowledge, this is the highest average power obtained from a ${{\rm WTe}_2}\mbox{-}{\rm SA}$WTe2-SA-based fiber laser. We further amplified the output power to 5.60 W with an all-fiber thulium-doped double-cladding fiber amplifier. Our result indicates that ${{\rm WTe}_2}\mbox{-}{\rm SA}$WTe2-SA could be an excellent candidate for a high-power fiber laser system.

Fundamental and harmonic mode-locked h-shaped pulse generation using a figure-of-9 thulium-doped fiber laser.

Herein, we present a fundamental and harmonic mode-locked figure-of-9 thulium-doped fiber laser using a nonlinear amplifying loop mirror. The generated fundamental mode-locked h-shaped pulse is centered at 1889 nm with an average output power reaching 282 mW and a pulse energy up to 1.23 µJ, which is the highest power and pulse energy of an h-shaped pulse. In the harmonic mode-locked regime, up to the 8th harmonic h-shaped pulse is obtained. The detailed characteristics of the h-shaped pulse are discussed. The proposed study shows that the figure-of-9 fiber laser can generate h-shaped pulses and also allows the generation of nanosecond pulses with a µJ-level pulse energy.

A Structure-Adaptive Hybrid RBF-BP Classifier with an Optimized Learning Strategy.

This paper presents a structure-adaptive hybrid RBF-BP (SAHRBF-BP) classifier with an optimized learning strategy. SAHRBF-BP is composed of a structure-adaptive RBF network and a BP network of cascade, where the number of RBF hidden nodes is adjusted adaptively according to the distribution of sample space, the adaptive RBF network is used for nonlinear kernel mapping and the BP network is used for nonlinear classification. The optimized learning strategy is as follows: firstly, a potential function is introduced into training sample space to adaptively determine the number of initial RBF hidden nodes and node parameters, and a form of heterogeneous samples repulsive force is designed to further optimize each generated RBF hidden node parameters, the optimized structure-adaptive RBF network is used for adaptively nonlinear mapping the sample space; then, according to the number of adaptively generated RBF hidden nodes, the number of subsequent BP input nodes can be determined, and the overall SAHRBF-BP classifier is built up; finally, different training sample sets are used to train the BP network parameters in SAHRBF-BP. Compared with other algorithms applied to different data sets, experiments show the superiority of SAHRBF-BP. Especially on most low dimensional and large number of data sets, the classification performance of SAHRBF-BP outperforms other training SLFNs algorithms.

A Feature Point Matching Based on Spatial Order Constraints Bilateral-Neighbor Vote.

Feature point matching is a fundamental and challenging problem in many computer vision applications. In this paper, a robust feature point matching algorithm named spatial order constraints bilateral-neighbor vote (SOCBV) is proposed to remove outliers for a set of matches (including outliers) between two images. A directed k nearest neighbor (knn) graph of match sets is generated, and the problem of feature point matching is formulated as a binary discrimination problem. In the discrimination process, the class labeled matrix is built via the spatial order constraints defined on the edges that connect a point to its knn. Then, the posterior inlier class probability of each match is estimated with the knn density estimation and spatial order constraints. The vote of each match is determined by averaging all posterior class probabilities that originate from its associative inliers set and is used for removing outliers. The algorithm iteratively removes outliers from the directed graph and recomputes the votes until the stopping condition is satisfied. Compared with other popular algorithms, such as RANSAC, RSOC, GTM, SOC and WGTM, experiments under various testing data sets demonstrate strong robustness for the proposed algorithm.

Inverse consistent non-rigid image registration based on robust point set matching.

BACKGROUND: Robust point matching (RPM) has been extensively used in non-rigid registration of images to robustly register two sets of image points. However, except for the location at control points, RPM cannot estimate the consistent correspondence between two images because RPM is a unidirectional image matching approach. Therefore, it is an important issue to make an improvement in image registration based on RPM. METHODS: In our work, a consistent image registration approach based on the point sets matching is proposed to incorporate the property of inverse consistency and improve registration accuracy. Instead of only estimating the forward transformation between the source point sets and the target point sets in state-of-the-art RPM algorithms, the forward and backward transformations between two point sets are estimated concurrently in our algorithm. The inverse consistency constraints are introduced to the cost function of RPM and the fuzzy correspondences between two point sets are estimated based on both the forward and backward transformations simultaneously. A modified consistent landmark thin-plate spline registration is discussed in detail to find the forward and backward transformations during the optimization of RPM. The similarity of image content is also incorporated into point matching in order to improve image matching. RESULTS: Synthetic data sets, medical images are employed to demonstrate and validate the performance of our approach. The inverse consistent errors of our algorithm are smaller than RPM. Especially, the topology of transformations is preserved well for our algorithm for the large deformation between point sets. Moreover, the distance errors of our algorithm are similar to that of RPM, and they maintain a downward trend as whole, which demonstrates the convergence of our algorithm. The registration errors for image registrations are evaluated also. Again, our algorithm achieves the lower registration errors in same iteration number. The determinant of the Jacobian matrix of the deformation field is used to analyse the smoothness of the forward and backward transformations. The forward and backward transformations estimated by our algorithm are smooth for small deformation. For registration of lung slices and individual brain slices, large or small determinant of the Jacobian matrix of the deformation fields are observed. CONCLUSIONS: Results indicate the improvement of the proposed algorithm in bi-directional image registration and the decrease of the inverse consistent errors of the forward and the reverse transformations between two images.

Elastic image registration using hierarchical spatially based mean shift.

In this paper, a novel estimation technique for corresponding points using a hierarchical, spatially based mean shift algorithm is proposed. We proposed a spatially based probability estimation using different spatial masks. For a given point on reference image, its corresponding register point is found along the search trajectory generated by optimizing Bhattacharyya coefficient between two windows centered at the points on the register and reference images. The outliers are further eliminated by analyzing statistical information on the displacements of the candidate register points. Experiments on various monomodal medical images show that the proposed method is feasible and fast.

[Medical image registration based on hamming window filtering and particle swarm optimizaton].

Local maxima in multimodality image registration based on mutual information is discussed in this paper. Particle swarm optimization (PSO) and filter preprocessing based on hamming window is used to search the registration parameters. Simulations have been done to illustrate that after low-pass filter preprocessing local maxima is eliminated to a great extent. In most case the global maxima can be found by PSO. Simulations illustrate the efficiency and accuracy of this method in registration strategy.