Cancer staging detection is important for clinician to assess the patients’ status and make optimal therapy decision. In this study, the machine learning algorithm based on principal component analysis (PCA) and support vector machine (SVM) was combined with urine surface-enhanced Raman scattering (SERS) spectroscopy for improving the identification of colorectal cancer (CRC) at early and advanced stages. Two discriminant methods, linear discriminant analysis (LDA) and SVM were compared, and the results indicated that the diagnostic accuracy of SVM (93.65%) was superior to that of LDA (80.95%). This exploratory study demonstrated the great promise of urine SERS spectra along with PCA-SVM for facilitating more accurate detection of CRC at different stages.

Multi-object tracking (MOT) techniques have been increasingly applied in a diverse range of tasks. Unmanned aerial vehicle (UAV) is one of its typical application scenarios. Due to the scene complexity and the low resolution of moving targets in UAV applications, it is difficult to extract target features and identify them. In order to solve this problem, we propose a new re-identification (re-ID) network to extract association features for tracking in the association stage. Moreover, in order to reduce the complexity of detection model, we perform the lightweight optimization for it. Experimental results show that the proposed re-ID network can effectively reduce the number of identity switches, and surpass current state-of-the-art algorithms. In the meantime, the optimized detector can increase the speed by 27% owing to its lightweight design, which enables it to further meet the requirements of UAV tracking tasks.

The bistratified lobula giant type 1 (BLG1) neuron is an identified looming-sensitive neuron in crab’s visual brain that demonstrates special sensitivity to diving targets, or descending approaching motions. In this paper, a novel neural model is proposed to shape such unique selectivity through incorporating a bio-plausible feedforward contrast inhibition synapse and a radially extending spatial enhancement distribution. Herein the synaptic connections and neuronal functions of this model are placed within a framework for matching and describing underlying biological findings. The systematic and comparative experiments have validated the proposed computational model that reconciles with the characteristics of BLG1 neurons in crab.

Most binary networks apply full precision convolution at the first layer. Changing the first layer to the binary convolution will result in a significant loss of accuracy. In this paper, we propose a new approach to solve this problem by widening the data channel to reduce the information loss of the first convolutional input through the sign function. In addition, widening the channel increases the computation of the first convolution layer, and the problem is solved by using group convolution. The experimental results show that the accuracy of applying this paper's method to state-of-the-art (SOTA) binarization method is significantly improved, proving that this paper's method is effective and feasible.

Manufacturing and agricultural industries use manual methods to count materials. This leads to low accuracy and inefficiency. This paper proposes a secondary counting method that combines main and differential counting. The area-fill identification algorithm is applied to mark the counted materials. To verify the effectiveness of the proposed counting algorithm, numbers of countings are conducted for different materials, such as the screws, hole gaskets, beans, jujube, etc. The results show that the counting accuracy reaches 98% for materials with size of 2—20 mm. The method has delivered a high-efficiency and high-accuracy automatic intelligent counting, with a wide range of application prospects and reference value.

The performance of decoding algorithm is one of the important influential factors to determine the communication quality of optical camera communication (OCC) system. In this paper, we first propose a decoding algorithm with adaptive thresholding based on the captured pixel values under an ideal environment, and then we further propose a decoding algorithm with multiple features, which is more suitable under the existence of the interference of light sources. The algorithm firstly determines the light-emitting diode (LED) array profile information by removing the interfering light sources through geometric features, and then identifies the LED state by calculating two grayscale features, the average gray ratio (AGR) and the gradient radial inwardness (GRI) of the LEDs, and finally obtains the LED state matrix. The experimental results show that the bit error ratio (BER) of the decoding algorithm with multiple features decreases from 1×10-2 to 5×10-4 at 80 m.

In this paper, we propose a single-port dual-beam leaky-wave antenna (LWA) in the terahertz (THz) band based on a composite spoof surface plasmon polariton (SSPP) waveguide. The antenna can generate three independent transmission channels by exciting two independent modes inherent to hole and groove structures, respectively. By periodic modulation of the hole and groove structures, we achieve dual-beam scanning through a broad radiation angle using only the -1st space harmonics of the two modes, hence avoiding the instability of the -2rd space harmonic. Within the operating frequency range of 0.62—0.85 THz, the gain ranges from 13.5 dBi to 17 dBi for the backward beam, and from 6 dBi to 11.8 dBi for the forward beam. The antenna can accomplish continuous backward beam through broadside to forward beam scanning with a total scanning range of 116° and an average efficiency of about 92%. The antenna exhibits a great potential in the design of multi-transceiver radar system in the THz band and multi-beam LWAs.

Metalenses are two-dimensional planar metamaterial lenses, which have the advantages of high efficiency and easy integration. However, most metalenses cannot modulate the light intensity, which limits their applications. To deal with it, taking advantage of flexible regulation of the beam amplitude and phase by the metalens, the geometric phase method is selected to design the dual-function metalens. It can effectively eliminate chromatic aberration in a visible light band from 535 nm to 600 nm and achieve amplitude modulation. After transmitting the metalens, the amplitudes of the beam respectively turn into 0.2 and 0.9. In this way, the amount of transmission of metalens in the preset band can be quantitatively controlled. According to the distribution characteristics of light diffraction intensity, the metalens designed can play a dual modulation role of achromatism and interference double-beam equilibrium in the paper, to meet the needs of miniaturization and integration of the optical system. The achromatic and amplitude-modulated metalens will have great application potential in optical holographic imaging and super-resolution focusing.

In this paper, we have evaluated a bidirectional wavelength division multiplexing passive optical network (WDM-PON) employing intensity modulated/direct detection optical orthogonal frequency division multiplexing (IM/DD-OFDM). The proposed system employs 100 Gbit/s 16 quadrature amplitude modulation (16-QAM) downstream and 5 Gbit/s on-off keying (OOK) upstream wavelengths, respectively. The proposed system is considered low-cost as non-coherent IM/DD OFDM technology and a simple reflective semiconductor optical amplifier (RSOA) colorless transmitter are employed and no dispersion compensating fiber (DCF) is needed. Based on the bit error rate (BER) results of WDM signals, the proposed WDM-PON system can achieve up to 1.6 Tbit/s (100 Gbit/s/λ × 16 wavelengths) downstream transmission over a 30 km single mode fiber (SMF).

It's common to use the method of continuous spectroscopy in water quality testing. But there're some problems with it. For example, the scanning results have a large number of nonlinear signals, and the covariance between variables is serious, which can lead to a decrease in the model prediction accuracy. In this paper, the standard solutions of nitrate nitrogen (NO3-N) and nitrite nitrogen (NO2-N) were used as the subject to be tested, and the data of the scanned waves and absorbance were obtained by use of spectral detector. The data were processed by noise reduction first and then the random forest (RF) algorithm was adopted to establish the regression relationship between concentration and absorbance. For comparison, partial least squares (PLS) and support vector machine (SVM) algorithm models were also established. For the same given data, the three reverse models can make the projection of the concentration respectively. The experimental results show that the RF algorithm predicts NO2-N concentrations significantly better than the SVM algorithm and PLS algorithm. This proves that the RF algorithm has good prediction ability in spectral water quality detection because of its high model accuracy and better adaptability, which could be a reference for similar research on continuous spectral water quality online detection.