Sea-Sky Line and its Nearby Ships Detection Based on the Motion Attitude of Visible Light Sensors.

In the maritime scene, visible light sensors installed on ships have difficulty accurately detecting the sea-sky line (SSL) and its nearby ships due to complex environments and six-degrees-of-freedom movement. Aimed at this problem, this paper combines the camera and inertial sensor data, and proposes a novel maritime target detection algorithm based on camera motion attitude. The algorithm mainly includes three steps, namely, SSL estimation, SSL detection, and target saliency detection. Firstly, we constructed the camera motion attitude model by analyzing the camera's six-degrees-of-freedom motion at sea, estimated the candidate region (CR) of the SSL, then applied the improved edge detection algorithm and the straight-line fitting algorithm to extract the optimal SSL in the CR. Finally, in the region of ship detection (ROSD), an improved visual saliency detection algorithm was applied to extract the target ships. In the experiment, we constructed SSL and its nearby ship detection dataset that matches the camera's motion attitude data by real ship shooting, and verified the effectiveness of each model in the algorithm through comparative experiments. Experimental results show that compared with the other maritime target detection algorithm, the proposed algorithm achieves a higher detection accuracy in the detection of the SSL and its nearby ships, and provides reliable technical support for the visual development of unmanned ships.

Sensitive Detection of Small Particles in Fluids Using Optical Fiber Tip with Dielectrophoresis.

This work presents using a tapered fiber tip coated with thin metallic film to detect small particles in water with high sensitivity. When an AC voltage applied to the Ti/Al coated fiber tip and indium tin oxide (ITO) substrate, a gradient electric field at the fiber tip induced attractive/repulsive force to suspended small particles due to the frequency-dependent dielectrophoresis (DEP) effect. Such DEP force greatly enhanced the concentration of the small particles near the tip. The increase of the local concentration also increased the scattering of surface plasmon wave near the fiber tip. Combined both DEP effect and scattering optical near-field, we show the detection limit of the concentration for 1.36 µm polystyrene beads can be down to 1 particle/mL. The detection limit of the Escherichia coli (E. coli) bacteria was 20 CFU/mL. The fiber tip sensor takes advantages of ultrasmall volume, label-free and simple detection system.

Determination of the effective index and thickness of biomolecular layer by Fano resonances in gold nanogrid array.

We present an accurate method to determine the effective refractive index and thickness of biomolecular layer by using Fano resonance modes in dual-period gold nanogrid arrays. The effective refractive index changes along the x and y directions are simultaneously measured and obtained by using a modified dispersion relation. The thickness of the surface layer is calculated by a three-layer waveguide equation without any fitting parameters. The accuracy of the proposed method is verified by comparing the results with the known coated dielectric layer and self-assembly layers. The applications of this method and nanogrid chips for determining the thickness and surface concentration of antigen/antibody interactions are demonstrated.

Increased detection sensitivity of surface plasmon sensors using oblique induced resonant coupling.

Increased detection sensitivity was achieved by adjusting the incident angle on periodic gold nanostructures that induced a resonant coupling between surface and substrate surface plasmon modes. For 500 nm-period gold nanoslits, a small incident angle, 7°, resulted in 2.64 times narrower linewidth and a 1.8 times increase in the figure of merit as compared to normal incidence. Furthermore, the intensity sensitivity was increased 4.5 times due to the change in the resonant coupling and redshift of the surface plasmon mode.

A Novel Reformed Reduced Kernel Extreme Learning Machine with RELIEF-F for Classification.

With the exponential growth of the Internet population, scientists and researchers face the large-scale data for processing. However, the traditional algorithms, due to their complex computation, are not suitable for the large-scale data, although they play a vital role in dealing with large-scale data for classification and regression. One of these variants, which is called Reduced Kernel Extreme Learning Machine (Reduced-KELM), is widely used in the classification task and attracts attention from researchers due to its superior performance. However, it still has limitations, such as instability of prediction because of the random selection and the redundant training samples and features because of large-scaled input data. This study proposes a novel model called Reformed Reduced Kernel Extreme Learning Machine with RELIEF-F (R-RKELM) for human activity recognition. RELIEF-F is applied to discard the attributes of samples with the negative values in the weights. A new sample selection approach, which is used to further reduce training samples and to replace the random selection part of Reduced-KELM, solves the unstable classification problem in the conventional Reduced-KELM and computation complexity problem. According to experimental results and statistical analysis, our proposed model obtains the best classification performances for human activity data sets than those of the baseline model, with an accuracy of 92.87 % for HAPT, 92.81 % for HARUS, and 86.92 % for Smartphone, respectively.

Multilabel Video Classification Model of Navigation Mark's Lights Based on Deep Learning.

At night, buoys and other navigation marks disappear to be replaced by fixed or flashing lights. Navigation marks are seen as a set of lights in various colors rather than their familiar outline. Deciphering that the meaning of the lights is a burden to navigators, it is also a new challenging research direction of intelligent sensing of navigation environment. The study studied initiatively the intelligent recognition of lights on navigation marks at night based on multilabel video classification methods. To capture effectively the characteristics of navigation mark's lights, including both color and flashing phase, three different multilabel classification models based on binary relevance, label power set, and adapted algorithm were investigated and compared. According to the experiment's results performed on a data set with 8000 minutes video, the model based on binary relevance, named NMLNet, has highest accuracy about 99.23% to classify 9 types of navigation mark's lights. It also has the fastest computation speed with least network parameters. In the NMLNet, there are two branches for the classifications of color and flashing, respectively, and for the flashing classification, an improved MobileNet-v2 was used to capture the brightness characteristic of lights in each video frame, and an LSTM is used to capture the temporal dynamics of lights. Aiming to run on mobile devices on vessel, the MobileNet-v2 was used as backbone, and with the improvement of spatial attention mechanism, it achieved the accuracy near Resnet-50 while keeping its high speed.

Spectral contrast imaging method for mapping transmission surface plasmon images in metallic nanostructures.

We propose a spectral contrast method to map the transmission images of surface plasmon resonance (SPR) in metallic nanostructures. Comparing the intensities between two neighboring wavelength bands near the SPR wavelength, the signal-to-noise ratio for biosensing applications obtained using the proposed method is found to be ten times higher than that obtained by conventional intensity analysis and 1.6 times better than that obtained by peak-wavelength fitting. The dynamic range and linearity of the refractive index are comparable to the peak-wavelength shift measurement. Based on the detection method, a spectral modulation system for the optical microscope is developed, combined with a gold-capped nanowire array, to measure the biointeractions in microfluidic devices. The experimental results show that the proposed method obtained multiple detections with a detection limit of 1.04 × 10-5 refractive index units. Two types of analysis methods for SPR images are used to study the protein-antibody interactions. The region-of-interest analysis supports multiplexing detections in a compact microfluidic sensor. The effective pixel analysis eliminates low-response pixels and enhances the signal-to-noise ratios for sensitive label-free detection.

Resonant position tracking method for smartphone-based surface plasmon sensor.

We propose a position-sensitive measurement method for tracking resonant signals of surface plasmon resonance (SPR) in periodic metallic nanostructures. Compared with conventional measurement methods, such as wavelength interrogation, intensity interrogation and spectral integration, this approach provides superior noise reduction and simple calculation process. Experimental results show that the limit of detection reaches to 5.88 × 10-6 RIU by using a portable spectrometer with a spectral resolution of 0.4 nm. The relationship between shot noise and signal noise was theoretically compared. The superior noise reduction of the resonant position tracking method is useful for the smartphone-based SPR measurement. The protein-antibody interactions using the smartphone and gold nanoslit arrays as the SPR sensor are demonstrated. It verifies that a smartphone can be used for sensitive measurement of binding-kinetics of the proteins.

Enhancing Surface Sensing Sensitivity of Metallic Nanostructures using Blue-Shifted Surface Plasmon Mode and Fano Resonance.

Improving surface sensitivities of nanostructure-based plasmonic sensors is an important issue to be addressed. Among the SPR measurements, the wavelength interrogation is commonly utilized. We proposed using blue-shifted surface plasmon mode and Fano resonance, caused by the coupling of a cavity mode (angle-independent) and the surface plasmon mode (angle-dependent) in a long-periodicity silver nanoslit array, to increase surface (wavelength) sensitivities of metallic nanostructures. It results in an improvement by at least a factor of 4 in the spectral shift as compared to sensors operated under normal incidence. The improved surface sensitivity was attributed to a high refractive index sensitivity and the decrease of plasmonic evanescent field caused by two effects, the Fano coupling and the blue-shifted resonance. These concepts can enhance the sensing capability and be applicable to various metallic nanostructures with periodicities.

Chip-based digital surface plasmon resonance sensing platform for ultrasensitive biomolecular detection.

A chip-based ultrasensitive surface plasmon resonance (SPR) sensor in a checkerboard nanostructure on plastic substrates is presented for digital detection. The sensing elements on the checkerboard are composed of silver-capped nanoslit arrays, which were fabricated using the thermal-embossing nanoimprint method, to meet the demand for low-cost and rapid fabrication. Sharp Fano resonances in the optimized nanoslit arrays provide high-intensity sensitivities (20,000% per refractive index unit), with an element size of 12.5µm. The polarization-dependent transmission in the checkerboard pattern produces optical isolation between sensing elements and results in a crosstalk lower than 1%. Protein-antibody experiments demonstrated that the digital detection limit was up to 1pg/mL, which is approximately 1000 times lower than that of conventional analog detection. For a 140µm×140µm checkerboard pattern, the dynamic range was approximately 100 times higher than that of conventional surface plasmon resonance measurements. This new digital detection method is very useful for detecting ultralow concentrations of analytes with a nonuniform distribution on the sensor surface.

Highly Sensitive Aluminum-Based Biosensors using Tailorable Fano Resonances in Capped Nanostructures.

Metallic nanostructure-based surface plasmon sensors are capable of real-time, label-free, and multiplexed detections for chemical and biomedical applications. Recently, the studies of aluminum-based biosensors have attracted a large attention because aluminum is a more cost-effective metal and relatively stable. However, the intrinsic properties of aluminum, having a large imaginary part of the dielectric function and a longer evanescent length, limit its sensing capability. Here we show that capped aluminum nanoslits fabricated on plastic films using hot embossing lithography can provide tailorable Fano resonances. Changing height of nanostructures and deposited metal film thickness modulated the transmission spectrum, which varied from Wood's anomaly-dominant resonance, asymmetric Fano profile to surface plasmon-dominant resonance. For biolayer detections, the maximum surface sensitivity occurred at the dip of asymmetric Fano profile. The optimal Fano factor was close to -1.3. The wavelength and intensity sensitivities for surface thickness were up to 2.58 nm/nm and 90%/nm, respectively. The limit of detection (LOD) of thickness reached 0.018 nm. We attributed the enhanced surface sensitivity for capped aluminum nanoslits to a reduced evanescent length and sharp slope of the asymmetric Fano profile. The protein-protein interaction experiments verified the high sensitivity of capped nanostructures. The LOD was down to 236 fg/mL.

Enhancing Surface Sensitivity of Nanostructure-Based Aluminum Sensors Using Capped Dielectric Layers.

The studies of nanostructure-based aluminum sensors have attracted huge attention because aluminum is a more cost-effective plasmonic material. However, the intrinsic properties of the aluminum metal, having a large imaginary part of the dielectric function and a longer electromagnetic field decay length and problems of poor long-term chemical stability, limit the surface-sensing capability and applicability of nanostructures. We propose the combination of capped aluminum nanoslits and a thin-capped dielectric layer to overcome these limitations. We show that the dielectric layer can positively enhance the wavelength sensitivities of the Wood's anomaly-dominant resonance and asymmetric Fano resonance in capped aluminum nanoslits. The maximum improvement can be reached by a factor of 3.5. Besides, there is an optimal layer thickness for the surface sensitivity because of the trade-off relationship between the refractive index sensitivity and decay length. We attribute the enhanced surface sensitivity to a reduced evanescent length, which is confirmed by the finite difference time-domain calculations. The protein-protein interaction experiments verify the high-surface sensitivity of the structures, and a limit of quantification (LOQ) of 1 pg/mL anti-bovine serum albumin is achieved. Such low-cost, highly sensitive aluminum-based nanostructures can benefit various sensing applications.

Enhancing the Surface Sensitivity of Metallic Nanostructures Using Oblique-Angle-Induced Fano Resonances.

Surface sensitivity is an important factor that determines the minimum amount of biomolecules detected by surface plasmon resonance (SPR) sensors. We propose the use of oblique-angle-induced Fano resonances caused by two-mode coupling or three-mode coupling between the localized SPR mode and long-range surface plasmon polariton modes to increase the surface sensitivities of silver capped nanoslits. The results indicate that the coupled resonance between the split SPR (-kSPR) and cavity modes (two-mode coupling) has a high wavelength sensitivity for small-angle incidence (2°) due to its short decay length. Additionally, three-mode coupling between the split SPR (-kSPR), substrate (+kSub) and cavity modes has a high intensity sensitivity for large-angle incidence due to its short decay length, large resonance slope and enhanced transmission intensity. Compared to the wavelength measurement, the intensity measurement has a lower detectable (surface) concentration below 1 ng/ml (0.14 pg/mm(2)) and is reduced by at least 3 orders of magnitude. In addition, based on the calibration curve and current system noise, a theoretical detection limit of 2.73 pg/ml (0.38 fg/mm(2)) can be achieved. Such a surface concentration is close to that of prism-based SPR with phase measurement (0.1-0.2 fg/mm(2) under a phase shift of 5 mdeg).

Spectral and mode properties of surface plasmon polariton waveguides studied by near-field excitation and leakage-mode radiation measurement.

We present a method to couple surface plasmon polariton (SPP) guiding mode into dielectric-loaded SPP waveguide (DLSPPW) devices with spectral and mode selectivity. The method combined a transmission-mode near-field spectroscopy to excite the SPP mode and a leakage radiation optical microscope for direct visualization. By using a near-field fiber tip, incident photons with different wavelengths were converted into SPPs at the metal/dielectric interface. Real-time SPP radiation images were taken through leakage radiation images. The wavelength-dependent propagation lengths for silver- and gold-based DLSPPWs were measured and compared. It confirms that silver-based SPP has a propagation length longer than a gold-based one by 1.25, 1.38, and 1.52 times for red, green, and blue photons. The resonant coupling as a function of wavelength in dual DLSPPWs was measured. The coupling lengths measured from leakage radiation images were in good agreement with finite-difference time domain simulations. In addition, the propagation profile due to multi-SPP modes interference was studied by changing position of the fiber tip. In a multimode DLSPPW, SPP was split into two branches with a gap of 2.237 µm when the tip was at the center of the waveguide. It became a zigzag profile when the SPP was excited at the corner of the waveguide.