This PDF file contains the front matter associated with SPIE Proceedings Volume 12252, including the Title Page, Copyright information, Table of Contents, and Conference Committee listings.

During the flight of the aircraft, it is necessary to monitor the maximum principal strain of the aircraft's own cabin or tank. For the acquisition of the maximum principal strain, the fiber Bragg grating strain sensor is often pasted on the surface of the object to be measured in the mode of three-directional strain rosettes. The relevant measurement results are brought into the calculation formula to obtain the maximum principal strain. Aiming at the current situation that when a large number of fiber Bragg grating strain sensors are pasted on a large structure, it cannot be calibrated and can only be calibrated with empirical values. According to the large error source introduced by the sensor in the actual mounting process, a method for obtaining high-consistency strain of the fiber grating strain sensor is presented, which greatly improves the strain measurement accuracy of the fiber Bragg grating strain sensor on large structures.

The accuracy of the metal oxide (MOS) gas sensor for mixed gas detection is reduced due to cross-sensitivity. This paper presents a new method for detecting binary mixed gas. This method preprocesses multi-channel signal data from gas sensor array using kernel principal component analysis (KPCA). Then the target gas is recognized by the support vector machine (SVM). The backpropagation artificial neural network (BP-ANN) was used to quantitative analysis of mixed gas. Genetic algorithm (GA) has the advantage of a global search to optimize the weight and threshold of the BP-ANN. The experimental samples of ethanol and acetone as the target gases were designed and realized to verify the proposed binary mixture gas detection method. Compared with Principal component analysis (PCA)for feature extraction, the recognition accuracy of gas species improved by 4.54%, reaching 98.27%. Compared with the concentration estimation method of the BP-ANN, the estimation accuracy is improved. The concentration estimation errors of ethanol and acetone increased by 1.72% and 0.94% respectively.

With the rapid development of the energy industry, the power industry has also begun to rise, especially the household meter system, which is being implemented now, so that the mining and approval of electricity meters are also increasing. Using BPNN algorithm, this paper studies the DN operation and maintenanceIMS, and discusses the functional requirements of DN operation and maintenance information management, as well as BPNN and design scheme; The statistical analysis of user and load control data in the DN operation and maintenance information management statistical database shows that the average load rate of DN lines of industrial users is as high as 81% and the utilization rate of power station interval is 23%; The average load rate of DN lines of users in administrative office and the utilization rate of power station interval are 62%. The confidentiality of user basic information, user load data sheet, line load and economic data sheet has reached more than 80%, and the line load operation efficiency is only 53%, which needs to be improved.

Non toxic and pollution-free cryogenic liquid propellant is widely used in the propulsion system of space vehicle. At the same time, higher requirements are put forward for the liquid level height monitoring sensor in liquid propellant. In order to judge whether the low-temperature capacitive liquid level sensor in the new generation launch vehicle meets the application requirements, a set of low-temperature liquid level test device is designed, and strict calibration test experiments are carried out before it is put into use. Therefore, liquid nitrogen is selected as the liquid medium, and the servo motor control system, grating ruler measurement system and data acquisition and processing system are used to simulate the sensor. The designed capacitance type liquid level sensor can meet the real-time calibration requirements of the low-temperature liquid level sensor through the low-temperature test.

With the widespread application of artificial intelligence technology in the military field, a distributed networked electronic warfare system consisting of a large number of confrontation nodes based on unmanned intelligent platforms is emerging. Due to the characteristics and advantages of intelligence, coordination, self-adaptation, and selforganization, distributed electronic warfare networks have become an important research direction in the current electronic warfare field1 . The research of combat system based on complex network provides a new perspective for the research of combat command and control system, which helps people to grasp the complexity of networked warfare as a whole, and to better understand its topological structure and corresponding dynamics feature. This paper analyzes from the characteristics of distributed electronic warfare network, introduces the HK (Holme-Kim) network model, and combines the operational characteristics of distributed electronic warfare system, selects HK (Holme-Kim) network model to build a complex network model of distributed warfare network, and analyzed its advantages.

The method of using the detection of temperature gradient is a new method to achieve turbulence observation. Most of the currently used temperature sensors are negative temperature coefficient nonlinear thermistors, but the degree of linearization of thermistor temperature sensors is an important factor affecting the accuracy of turbulence observation. In this paper, a low-cost two-stage linearization scheme for thermistor temperature sensors is proposed to address this problem. The first stage is a hardware linearization circuit based on the OPA2277 integrated operational amplifier. The second-stage linearization adopts the fitting between the first-stage output voltage and the real temperature based on the fourth-order fitting temperature calibration method. This linearization scheme trades off fast response to temperature and measurement accuracy. The linearization scheme based on this thermistor temperature sensor is in the temperature range of 0~32°C, and the maximum absolute temperature error is only 0.056°C, which is much smaller than the linearization scheme in the existing literature.

Conventional rigid metals and alloys are characterized by complex processing and high cost due to their high hardness and high melting point. Low-melting-point alloy has the advantages of easy processing and high conductivity due to its low melting point, which is very suitable for application in sensor electrodes. In this study, InSnBi electrode with a serpentine structure was fabricated by microfluidic technology. We simulated and compared electrodes with different structures, and studied the influence of manufacturing process on electrode performance. It was found that the interface of metal hydrogel was closely linked. And the InSnBi alloy was shown with good biocompatibility. The fabricated electrode has the potential to be applied for continuous glucose sensing.

In order to demodulate the voltage signal output by the eddy current detection sensor, an amplitude and phase demodulation circuit suitable for the low frequency range (20k) is designed, and the phase shift error is controlled within ±0.5%. The amplitude-phase demodulation circuit is used in a self-designed tunnel magnetoresistive sensor (TMR) probe based on magnetic shielding, and the detection of crack-type defects of different sizes on the surface of 6061 aluminum alloy plate is realized. The experimental results show that the amplitude and phase demodulation circuit can effectively demodulate the amplitude and phase information of the signal to be measured, and the phase information is hardly affected by the excitation current and is sensitive to the excitation frequency. The optimal excitation frequency for phase orientation in the experiment is 5kHz. The demodulated amplitude and phase information can characterize the crack defects of different lengths (10mm, 20mm, 30mm) and different depths (1-5mm) on the 6061 aluminum alloy plate. Whether it is the voltage amplitude or the phase result, when the defect depth is within 3mm, the output is approximately linearly related to the defect depth.

Ocean current measurement plays an important role in marine scientific research. Studies on the lateral line system show that it can accurately perceive the strength, direction, water pressure and other information of the surrounding water flow. Based on this biological principle, the bionic lateral line sensor is designed with PVDF, which is a polymer piezoelectric material with outstanding piezoelectric characteristics. The parameter and structure of the sensor are optimized by the computational fluid dynamics method. The theoretical model of the sensor is established by combining the hydrodynamics and piezoelectric effect. The experimental results show that the actual measurement value of bionic line sensor fits the theoretical model well. The flow velocity measurement range is 0.02m/s-0.5m/s, and the accuracy is less than 4%.

In order to effectively identify the respiration and heartbeat information from the radar echo signal, reduce the positioning error, and improve the detection accuracy of the ultra-wideband radar life detector. The WA-EMD denoising and reconstruction algorithm is proposed, and the peak-to-peak value, variance, standard deviation and distance error of the PE method are compared and analyzed. The results show that the WA-EMD algorithm can effectively separate the respiration and heartbeat from the preprocessed signal. The detection target error of the PE method is determined to be less than 5%, which verifies the feasibility of the detection distance of the PE algorithm.

In order to improve the signal-to-noise ratio of ultra-wideband radar life detection echo signals, analyze the causes and types of ultra-wideband radar signal noise, build echo signal models in real measurement environments, and study various mechanisms for dealing with noise. Taking a human body with vital signs as an example, the radar signal echoes at a distance of 1m, 2m and 4m from the radar are tested, the radar signal waves are obtained, and the noise is targeted according to the type of clutter. The results show that the useless signal in the original echo matrix signal is reduced and the signal-to-noise ratio is improved. The local normalization method and automatic gain control method are used to automatically enhance the processed signal strengthened before.

For the traditional sliding mode observer usually chooses the sign switching function as the sliding mode surface, it is not continuous on the sliding mode surface, leading to serious chattering. In this paper, the sine function is selected as the sliding mode surface, thereby reducing the chattering of the system. A new phase-locked loop with built-in wave limiter is added to suppress the influence of the characteristic frequency disturbance signal in the identification of the back EMF on the rotor position. The simulation results show that the algorithm can effectively weaken the synovial observer chattering on the rotor position identification. The influence of the results verifies the correctness of the theoretical analysis in this paper and the effectiveness of the proposed strategy.

Based on the characteristics of offshore wind farms and the practical cases, an applicable design scheme of automatic fire alarm system is introduced, which includes the selection and setting of fire detectors, linkage control of associated fire protection systems, arrangement of fire telephones, system power supply and wiring, etc. In addition to its compliance with the requirements for unattended operation and land⁃based remote monitoring of the offshore wind farm, the design can also ensure the stability and reliability of the system, so as to guarantee the safe operation of the offshore wind farm.

As perfumes, cosmetics, and other relevant products with fragrances become increasingly popular in the market, some health problems caused by some compounds in these fragrance products are also raised, such as allergic contact dermatitis. What’s more, since it is legal for companies to label the ingredients merely as “Fragrance” for trade secret protection, there is an even higher risk towards customers’ health, such as cancers and some chronic diseases. This article aims to survey some current analysis techniques, including chromatography, spectroscopy, and electronic nose, utilized in fragrance identification and suggestions of their application in different organizations. The characteristics, working mechanism, previous work about their application in fragrance analysis, etc. are introduced and compared in this work. Although gas chromatography-mass spectrometry (GC-MS) is the most popular and commonly used analytical technique used by most organizations, it does not have complete advantages compared to other developed methods. Capillary electrophoresis (CE) is more environmentally friendly and efficient, and Raman spectroscopy (RS) is more accurate and cost-effective. These two methods are more promising for flavor identification when enterprises and governments conduct quality supervision in the future. Electric nose, with a more sensitive sensor but less precision in identifying fragrance compounds, is more suitable for classification and product counterfeiting.

Accurate and convenient detection of human glucose level is an effective method for early diagnosis of diabetes and prevention of complications. The flexible electrochemical glucose electrode with low cost, fast responsiveness, good stability, reliability, and high sensitivity has attracted much attention in the monitoring of glucose concentration. The preparation of a conductive layer with catalytic activity on a flexible substrate is the key to make a flexible glucose electrode. Here, we have successfully prepared graphene composite materials sintered with copper nanoparticles on a polyimide film by direct laser writing, and fabricated a flexible non-enzymatic glucose electrode using laser-ablated graphene (LAG) as a conductive electrode. The Cu/LAG electrode exhibits high catalytic activity for glucose. The glucose electrode shows a linear relationship with the sensitivity of 0.708 mA mM^-1 cm^-2 in 0.05-1.5 mM. In addition, The Cu/LAG electrode also exhibits excellent response time(~2s). The electrode has good application potential and can be used to monitor glucose in human sweat.

At present, with the advent of 5g and big data technology era, the development of intelligent agriculture industry based on information technology has been applied, in which the Internet of things is widely used to enrich planting technology, optimize the traceability of agricultural products, and ensure the organic combination of information technology and intelligent agriculture. By analyzing the indicators of greenhouse monitoring, this paper puts forward the standard architecture scheme of Intelligent Greenhouse Based on Agricultural Internet of things: in the perception control layer, the wireless sensor module detects soil moisture, environmental temperature and humidity and illumination, and uses the relevant knowledge of data fusion to start the control equipment such as lighting, temperature control and gas concentration, In the network transmission layer, Siemens S7-200 smart PLC and stm32f030c8t6 constitute the dual control core, sm3390b sensor is used to collect and monitor data, and the data hub between system layers is established; The terminal application layer adopts the visual man-machine interface of weiluntong, which can view the environmental monitoring data at any time and place. The simulation results show that the system has low cost but high automation level, which is of guiding significance to further improve the development of agricultural planting in China in the direction of intelligence.

For the convolutional neural network in the process of image segmentation, there are problems such as loss of detailed information and global features of the image, and the segmentation results of the full convolutional neural network in the image segmentation process are not refined and lack spatial consistency. This paper proposes a liver tumor segmentation method based on Multiple Supervised Fully Convolutional Networks (MSFCN). In the framework of a fully convolutional network, a supervised side output layer is added to the convolutional layer to guide multi-scale feature learning, which can better capture the local and global features of the image. Experiments are carried out on the LiTS competition dataset, and the experimental results show that the proposed method can significantly improve the segmentation accuracy.

Aiming at the problems of poor adaptability and insufficient functions of existing exploration robots in different caves and dangerous areas, this paper proposes a multi-wheel drive binocular exploration robot based on ROS system, which uses laser radar sensor and binocular depth vision camera to collect environmental data. The Gmapping algorithm is used to realize the real-time positioning and accurate map building of the robot system. The CCD vision camera collects the data of the exploration area in real time, and transmits the data to the host computer for analysis and processing. Through the ROS open source system, it is convenient to control the robot and receive the mapping data, navigation data and exploration data of the robot system on the upper computer. Through experiments,Compared with the existing exploration system, the system has the characteristics of low hardware cost, easy mastery, easy modification, strong function expandability and the like; This project has a very high application value in the exploration of caves and dangerous areas.

Radio frequency (RF) fingerprint refers to the inevitable subtle defects in the hardware circuit of radio transmitter, which are reflected in the signal waveform transmitted by it. Therefore, the wireless device can be uniquely identified by analyzing the received signal waveform. Due to the advantages of multidimensional mapping of convolutional neural network (CNN), it can automatically extract signal features and transmitter nonlinear features, which can not be extracted by traditional low-dimensional algorithms. Therefore, it is very important to design a deep neural network with reasonable structure and high recognition accuracy. Taking the in phase and quadrature (I & Q) time domain signals of eight radiation sources of the same model as the input samples, the fine features of IQ time domain signals are extracted by designing six deep learning radiation source individual recognition algorithms with different structures to realize the identification of RF fingerprint. Besides, the change of wireless channel probably weakens the identification robustness of RF fingerprint. Thus, a deep transfer learning algorithm is proposed to cross train and test the I & Q data collected at different times, which solves the problem of poor robustness.

In this paper, a multi-target recognition and distance detection method based on PE-Kmeans++ and HOC algorithm is proposed and tested experimentally. Firstly, the radar echo signal is preprocessed, and the number and distance of human targets are estimated based on PE-Kmeans++ algorithm. Then the body surface vibration signals of different targets are extracted to detect the vital signs of single target respectively. The respiration signal and heartbeat signal are restored according to the frequency range of respiration and heartbeat, and the corresponding peak frequency is extracted as the respiration and heartbeat frequency by the algorithm of HOC-FFT. Experimental results show that the proposed algorithm can accurately identify multiple human targets in the environment without complex preprocessing of the original echo signal.

In this paper, 45 male faces and 45 female faces were collected to evaluate the texture of human face, gray level cooccurrence Matrix (GLCM) was used to extract the texture features of male and female faces, such as energy, contrast, correlation and entropy. T test was performed with SPSS17.0 software, and the results were compared and analyzed, the experimental results show that there are statistical differences between men and women in the four features. Based on the analysis of texture and pore depth, Pearson Correlation Coefficient is 0.9795, and the correlation between pore depth and texture is very strong.

To improve the effectiveness of the infrared image dehazing algorithm, we modify the DCP based on the observation of several outdoor haze-free images: in these images, pixels of most local patches no longer have low intensity but are close to high intensity. The transmission map is estimated through the MDCP, and the guided filter and CLAHE achieve enhancement. The haze infrared images can be recovered completely. Moreover, to thrive the infrared dehazing research, we introduce the IR Dense-haze dataset, which contains 13 pairs of synthetic haze infrared images and corresponding haze-free infrared images and six extra haze infrared images captured in natural scenes. Experiments on the IR Densehaze dataset show that our method achieves dominant performance. The average PSNR of the infrared image is 39.02% higher than the original DCP method, and the average SSIM is 28.65% higher than the original DCP method.

Xi'an is the capital city of Shaanxi Province, the core city of the Guanzhong Plain urban agglomeration, an important central city in western China approved by the State Council, and an important national scientific research, education and industrial base. So creating Chinese Shaanxi Xi’an Dialect talking head has significant research value. In addition, Chinese Shaanxi Xi'an Dialect is also a precious cultural heritage which has great research value not only in linguistic field but also in speech analysis. We decided to take up this challenge to develop a facial animation system for this dialect lip and tone synchronization. The static viseme could be displayed by a stationary human facial image. However when a phoneme was articulated, the movement of our vocal organs，such as tongue or lips are more like a dynamic process than a static state. So we introduce the concept of dynamic tongue model here, which represents the entire process of the vocal organ movement during the articulation of a certain phoneme. The dynamic viseme model is a combination of dominance and parameter value based on dominance grade. I have already created a dynamic viseme model which is a dynamic articulation model based on dominance classification concept of the visemes for the Chinese Shaanxi Xi’an Dialect. It can be used in speech assistant systems for hard-of-hearing children or the people who wants to learn this dialect and provided a method to be used in speech assistant systems for other languages.

Bayesian Network, which has been widely used for its significant advantages in causal inference, has great potential in product design. However, it is difficult to learn a reasonable network structure facing the problem with small data during the design process. In order to solve this problem, this paper introduces expert knowledge and Genetic Algorithm based on matrix coding into the process of Bayesian Network structure learning. Integrating the expert knowledge and data is a decent way to make up for the problem of insufficient data, and Genetic Algorithm is used to improve traditional structure learning algorithms, so as to obtain a more suitable structure conforming to the knowledge and data characteristics. The solutions illustrate that the Genetic Algorithm has some advantages compared with traditional structure learning method, and the use of expert knowledge can improve the rationality of the learned structure.

In acidic medium and under the catalysis of KBr, orange yellow 1 can be oxidized by KIO₃ and this causes the solution to fade. A good linear relationship is obtained between the fading degree of the solution and KIO₃ dosage. Hereby, a new method for the determination of KIO₃ content in edible salt by fading spectrophotometry is established. The experimental results show that maximum absorption wavelength of orange yellow 1 is 490 nm in acidic medium, the linear relationship between the KIO₃ mass concentration and the change value of absorbance is ΔΑ=-0.0402+0.0745C (μg/mL) when the mass concentration of KIO₃ is 0.6000~2.000 μg/mL, and the linear correlation coefficient is 0.9987. This method had been successfully applied to determinate KIO₃ in edible salt, and the results are satisfactory.

With the continuous development of science and technology, a variety of information technology tools have emerged in the field of traditional animal husbandry to assist in production, distribution, and sales, thereby creating more economic value. Among them, the application of Internet of Things technology has played a very significant role. It can effectively transform the production mode of animal husbandry into intensive management and systematic breeding, thus promoting the rapid development of the animal husbandry industry. This paper mainly outlines the RFID-based Internet of Things technology, combined with the current situation of China's animal husbandry information construction, to explore the application and strategy of the Internet of Things in the animal husbandry industry, so as to provide a reference for the development of industrial innovation.