In order to increase the accuracy classified by image texture to remove plenty of useless information in wood hyper-spectral images, the texture classification algorithms based on the multifractal theory were used in this paper. Firstly, the most representative ten bands were screened out using different Feature selection algorithms, and then the multi-fractal curves of the selected bands were obtained according to different function density images. The multi-fractal curves were averaged, which could represent the texture feature of certain sample. Finally, the curves were classified by the Support Vector Machine (SVM) and BP neural network classifier. The result shows that the bands screened by K-L divergence are superior to that screened by adaptive band selection (ABS), the image’s texture features extracted by multi-fractal algorithm are better than that extracted by the gray level co-occurrence matrix(GLCM), and the classification accuracy and classification speed of SVM have an advantage over that used BP neural network. It can be concluded that the integrating K-L divergence, multi-fractal with SVM algorithm can dramatically increase the recognition classification, which can reach 97.91% in our work.

In order to solve the problem of low precision detection for small and medium-sized targets in the real-time traffic sign recognition, a lightweight traffic sign detection network based on YOLOv3-tiny depthwise separable convolution is proposed. In this paper, a deep separable convolution module is constructed by using deep separable convolution instead of the common convolution to construction feature extraction network,the feature information of small and medium-sized targets is better extracted under the premise of ensuring the calculation amount.At the same time, the multi-scale feature fusion network is improved to improve the detection accuracy of small and medium-sized traffic signs. The h-swish activation function is used to reduce the image features lost due to the increase of the number of network layers, and the detection of multiple types of traffic signs is realized.The experimental results show that the algorithm effectively improves the detection of small and medium-sized traffic signs. Warining, mandatory and prohibitory traffic signs are detected on the verification set. The detection accuracy (AP) is 98.57%, 96.03% and 98.04% respectively. The average detection accuracy (mAP) was 97.54% and the detection speed was 201.5 f/s. The average accuracy was 14.01% higher than that of YOLOv3-tiny. The algorithm effectively improves the detection accuracy under the premise of ensuring low calculation of light network and good timeliness of detection.

Aiming at the problems of current commercial river crab quality safety, hard to distinguish between real and false Yangcheng Lake hairy crabs, and too large feature dimension of deep residual network extraction, a precise source tracing system of river crab based on improved deep residual network is proposed. The system consists of four parts: breed, detection, sale and traceability. In breed, detection and sale parts, the data of crab breed, detection and sale are stored in traceability database. The traceability part identifies whether there is traceable river in traceability database by identifying crab based on improved deep residual network. According to the result of identification, the data of crab breeding, testing and marketing are output. Finally, the precise traceability of each commercial crab from consumer to farm is realized. The improved crab identification technology of the deep residual network can reduced the extracted crab shell feature vector from 2 048 to 156 dimensions, the recognition time is reduced by 92%, and the recognition accuracy is 92.1%.

In order to realize distributed secure storage of digital images, an image encryption-secret sharing algorithm is proposed in this paper. The image encryption function is completed by a sequence of four random numbers generated by a two-dimensional cascade modulation coupling map (2D-CMCM), which composed of a logistic map and an ICMIC map, including row/column scrambling and diffusion. In the secret sharing phase of the encrypted image, the encrypted image is divided into five secret subgraphs using a (3, 5) threshold secret sharing algorithm based on Lagrangian polynomial interpolation. The decryptor can obtain the lossless reconstruction of the original plaintext image only by acquiring no less than 3 secret subgraphs and secret keys. In the safety analysis experiment, the proposed algorithm shows a high degree of anti-aggression. In the reversibility analysis experiment, the secret subgraph is confirmed to achieve the lossless reconstruction of the original plaintext image. In the execution efficiency test, the high execution efficiency of the proposed algorithm is also verified. The a series of experiments con firmed the high practicability of the proposed algorithm.

Liquid crystal achromatic polarization rotator is an optical element which can change the state of incident polarized light. In order to realize the transformation of polarized light in orthogonal direction and have good achromatic performance in wide band without moving polarizing components, an achromatic polarization rotator composed of three low-twisted nematic liquid crystal cells is proposed to control the distribution of liquid crystal molecules by applying voltage to control the state of the polarized light. TechWiz LCD 1D software is used for simulation. In the case of no voltage and parallel polarizer, the light leakage rate of the chromatic polarization rotator in the range of 450~650 nm is less than 0.01, and the light leakage rate in the range of visible light is less than 45%, which is lower than that of the rotator composed of a traditional twisted nematic liquid crystal cell. When the voltage is applied, the transmittance can reach more than 97.9%, and when the thickness of the liquid crystal cell changes between 2.2~2.5 μm, the achromatic characteristic is almost unaffected. This structure of achromatic polarization rotator has high ratio, good achromatic performance and fast response time which meets the requirements of polarization rotator devices.

Organic electroluminescent blue light devices still have some problems, such as low efficiency and unsatisfactory operating lifetime. In order to achieve high efficiency long life blue light devices, a novel blue fluorescent luminescent material was obtained by 13,13-dimethyl-13H-indeno[1,2-b]anthracene connect Di-p-tolyl-amine. By device optimization it can achieve a current efficiency of 10.2 cd/A and a flatter efficiency roll-off. The peak spectrum was 462 nm, CIE (0.15, 0.17). The operating lifetime of the devices were tested at a constant current density of 10 mA/cm2 at room temperature. The device with this material as a blue light material reached a lifetime of 329 h, which was nearly twice as long as the lifetime of the reference device BCzVBi (169 h). The results show that the material can realize high-efficiency long-lifetime blue organic electroluminescent devices, especially for potential application of high-performance white organic light-emitting diodes and full-color display devices.

A new polyhalogenated biphenyl compound 4-bromo-4′-(bromodifluoromethyl)-2,3,3′,5′-tetrafluoro-1,1′-biphenyl was synthesized from 2,3,3′,5′-tetrafluoro-1,1′-biphenyl by reacting with n-BuLi and CF2Br2 under ultra-low temperature. But, this reaction can be simplified to a one-step reaction by constructing two different functional groups -CF2Br and -Br into the molecular structure. As a general intermediate, this compound can be used to synthesize a variety of difluoromethyl ether bridge-bonded liquid crystal monomers, which provides a new method for the preparation of such liquid crystal monomers. In this paper, the difluoromethyl ether bridge-bonded liquid crystal monomer 3PYUQUF was synthesized by the above methods. The results show that the Δn values of the compound 3PYUQUF was 0.138 0, and the Δε value was 22.63.

In the human-computer interaction scenario, in order to realize the hardware universal design of key matrix, function software custom settings, and the number of keys can be expanded, the OLED display keys are studied, and the key matrix based on OLED display keys is designed. According to the driving principle of OLED display keys, the circuit schematic diagram of key matrix based on LPC1788 controller is designed. The design scheme based on hardware generalization and the working principle of each part of the circuit are proposed. Then, according to the control sequence and state transformation of OLED display keys, the firmware design of LPC1788 controller and the configuration software design of the upper computer of OLED display keys are carried out, which realizes the configurability of the software for displaying text and pictures. Finally, the design of OLED display key matrix is tested, and the universal hardware settings, key function definable and extensible design are realized. The design and implementation show that OLED display keys can be designed as key matrix in human-computer interaction scene. The key display resolution is 96×64 pixels, 180° perspective, dimming range is 0~250 cd/m2, high definition, high contrast and full screen display. Through flexible configuration of software, matrix page turning design can be realized, which expands to more key demand and saves structure space. The OLED display key matrix design based on LPC1788 is stable, reliable, clear, versatile, beautiful in appearance, friendly in man-machine interface, and can be widely used in man-machine interaction system for keyboard requirements of vehicle, ship and airborne.

A kind of privacy display solution is developed using dual cell stack up, which can achieved switchable function between share mode and privacy mode. First we build the lighting-field display model in theory by dual cell LCD device, then extract data require by dual image generate algorithm, so can get dual cell image by deep learning algorithm. Meanwhile we simulate to evaluate the privacy mode, and assemble the dual cell sample to verify the result. By calculating and comparing the difference between side view images and original image, the view angle can has obviously anti-peep effect if SSIM <0.2. The actual sample experiment is basically consistent with the above result. According to the experiment, can verify the dual cell image multiplication mode following the principle of lighting field theory. The view angle over 45° can get the chaos image and achieve privacy display condition.

The influence of concentration (mass fraction, C1) and concentration of acid radical ion (mass fraction, C2) of TFT - LCD developer on Halftone THK is researched, and the way to improve THK fluctuations is put forward. Firstly, the production conditions of lithography process are fixed, then the influence of THK with C1 and C2 is respectively tested using the single factor experimental method. Studies show that the changes of C1 and C2 will affect the developer’s development ability and lead to the changes of THK. THK with C1 and C2 are both linear negative correlation. Secondly, according to the influence degree of C1 and C2 changes on THK, the formula corresponding relations between the two is established. The C1 is adjusted through the system and then the C2 fluctuations are improved. Lastly, the development ability and the THK fluctuations are stable through putting the formula into the system which make channel related incidence reduced to below 0.03% from more than 0.20%.

TFT White Color Uniformity (WCU) and SiNx film thickness of each layer were tested based on the same test points and the correlation between them is analyzed. It is found that TFT WCU has a great correlation with GI and PVX2 layers rather than PVX1 layer with the thinnest THK and largest n value. On this basis, an improvement scheme to reduce the center value of GI Remain THK(to avoid the residue of GL layer) and PVX2 THK (to improve the uniformity in panel)at the same time is proposed, which can reduces the mean value of TFT WCU by about 0.000 7. After improvement, the maximum value can be decreased from 0.007 0 to 0.005 2 to satisfy the specification(maximum value ≤ 0.005 5). The effect of improvement is obvious.

Flicker is an important index of TFT-LCD image quality evaluation. Based on the mechanism of ΔVp and flicker, the influence of Vcom on flicker is studied, and the experiment results show that Vcom has a significant impact on flicker. With the increase of flicker gray level, the optimum flicker tends to decrease gradually and its corresponding Vcom for TN mode products decreases gradually, while ADS mode products increases gradually. Secondly, the article studied The impact of VGH and VGLon flicker is stadied, the results indicate that VGH and VGL have important impact on flicker. When flicker gray level is invariable, the VGH increase will not affect the optimal flicker, but the corresponding Vcom decreases gradually. While the VGL increase the optimal flicker decreases gradually, but the corresponding Vcom increases gradually. A quantitative analysis method is proposed for predicting flicker scales of different gray level and proved that it can be used to predict the flicker scale of different gray level, because the calculated flicker is consistent with measurement flicker value by FMA test method under VESA standard, and the linear correlation coefficients are above 0.95 for both TN mode and ADS mode products. These studies provide analytical methods and solutions for improving TFT-LCD flicker problem.