Micro LED technology， as a cutting-edge research field in next-generation display technology， has advantages such as high brightness， high contrast， and high energy efficiency. The development process of Micro LED technology has been reviewed， focusing on its technical difficulties and current progress. The technical difficulties include epitaxial structure design， size effects in chip preparation， full color issues， Micro LED system integration， and reliability research. The current progress covers sidewall effect suppression， full color scheme， massive transfer technology， nitride red light technology， color conversion， vertical stacking， CMOS and TFT driving， 3D integration， transparent display， and nano LED. Finally， an outlook on the future development direction of Micro LED technology has been proposed， including addressing technical challenges， promoting industrialization， and achieving wider applications.

Currently， the most widely used three-dimensional（3D） display technology often uses the binocular parallax and transient visual effect of the human eye to generate the three-dimensional impression. The lack of depth information will lead to a conflict in the convergence adjustment and discomfort for the viewer. A true 3D display technology， light field display reproduces light on an object's surface， offers a positive user experience， and it satisfies people's needs for future displays. In this paper， the principles and characteristics of various light field display technologies were combed， the development of light field display technologies at home and abroad were analyzed. Then， the research progress and the performance of representative prototypes were sorted out. Finally， the development of light field display in China was prospected. Despite China's light field exhibition getting off to a rather late start， it has earned some notoriety abroad thanks to the combined efforts of various academic organizations.

Using 200 mm silicon-based GaN epitaxial wafers， wafer-level bonding， thin film transferring and micro-size pixel preparation were studied， and the lighting and display functions of Micro LED microdisplay array were realized. Compared with traditional sapphire-based GaN epitaxial wafers， 200 mm silicon-based GaN epitaxial wafers were more compatible with silicon-based drive wafer sizes， which was conducive to low-cost batch production through wafer bonding process. At the same time， the difficulties and ideas of batch production were analyzed， and it was pointed out that the defect was a problem that needed to be solved urgently.

A broken wire detection system of prestressed concrete cylinder pipe （PCCP） pipeline combined with phase‑sensitive optical time domain reflectometry （Φ‑OTDR） and Mach‑Zehnder interferometer （MZI） which reduced the amount of data computation by 40% through inducing optimized in‑phase quadrature （IQ） demodulation algorithm was designed. The results of the experiment showed 20 kHz frequency response and ±2 m spatial positioning error over 20 km long fiber for single-end measurement which indicated the system could position broken wire incidents along long distance and distinguish characteristic frequencies of vibration incidents. The proposed method provided a reliable technical means for online structure health detection with full lifecycle coverage of PCCP pipeline.

Aiming at the problem of insufficient detection accuracy of complex beach small object tourists， a beach small object tourist detection method based on multi-layer feature map information fusion was proposed， and the context and feature map information was used to improve the detection rate of small object tourists. Firstly， GCSAM structure was proposed by combining the more comprehensive and effective GAM attention mechanism idea with CSP structure. The GCSAM structure was used to enhance the cross-latitude receptive field of YOLOv5 backbone. The backbone network was focused on small object feature learning. Secondly， the PANet structure was replaced by BIFPN structure in YOLOv5 network to complete transmission of feature information across layers at the neck network， more context was included in the feature map. Finally， the power transform was used to improve CIOU_Loss to Alpha-CIOU_Loss in YOLOv5 network， and the regression accuracy of prediction frame was effectively improved. Experimental results showed that in comparison with original YOLOv5 network， the precision was improved by 2.00%， the recall was improved by 5.33%， and the mean average precision was improved by 4.36% with real-time requirements. Moreover， this method had better robustness in the case of dense tourists， occlusion， and smaller objects.

A Sb2S3-based transistor was fabricated to simulate synaptic behavior under optoelectronic stimuli. Channel conductance of the transistor could be controlled by optoelectronic stimuli through the gate. Additionally， the device was able to simulate long-term potential， paired-pulse depression， Pavlov's dog learning and extinction， and spike-timing dependent plasticity.

In order to solve the practical problem between fast speed of optical switch and slow speed of drive control circuit ， a design method of driving and controlling circuit for large-scale optical switch array was presented. High level or low level amplitude of the drive and control circuit could be arbitrarily adjusted by using FPGA to control the digital to analog converter and high-speed switch， which was creative. The new design principle of large-scale optical switch array drive control circuit was introduced completely through the combination of overall design， circuit design and software design. Finally， the test showed that the switching speed of the control circuit was less than 2 ns.

With the development of display devices towards high resolution and high refresh rate， the amount of video image data was increasing exponentially， which brought challenges to the bandwidth resources of the display system. From the perspective of human visual system， combining the change of the gaze point position and the sampling characteristics of the human eye foveal， an adaptive foveal image transmission algorithm based on human eye gaze point was proposed to reduce the data transmission bandwidth. The experimental results showed when the viewing distance was 2 and the gaze point position was at the center of the image， the comprehensive compression ratio was 12.257， the data transmission clock frequency was 12.532 MHz， and the total amount of data transmitted was 8.158% of the original total. The total amount of data transmission was greatly reduced， and the pressure on data transmission bandwidth was effectively relieved， so that video images with higher resolution and refresh rate could be displayed under the same clock frequency.

Aiming at the problems of cumbersome calibration process and low calibration accuracy of lidar and visible light camera， a joint calibration method of lidar and visible light camera based on EPnP algorithm was proposed. Firstly， an image feature points extraction method based on prior information was proposed. The coordinates of feature points in visible light image were calculated accurately by using the color information and the edge contour information of the calibration plate； Secondly， the feature points in 3D point cloud were extracted accurately according to the size information and edge contour information of the calibration plate by using the feature points extraction method based on line fitting proposed in this paper； Finally， the matching feature points extracted from the synchronous visible light image and the lidar three-dimensional point cloud were used as the input of the EPnP algorithm to solve the external parameter matrix between the lidar and the visible light camera. The experimental results showed that， compared with the Autoware calibration toolbox widely used in the industry， the proposed method could achieve higher calibration accuracy and simplify the tedious calibration process.

IGZO TFT is a typical representative of metal oxide TFT. The basic characteristics of IGZO-TFT devices and the influencing factors of electrical instability were analyzed. In the study of influencing factors， a reasonable explanation was given for the dependence between the film performance of the active protective layer and the electrical properties of TFT. And the preparation method of the active protective layer was optimized through experimental verification， which could solve the problem of poor display of left shift of TFT switch threshold voltage caused by the density of the active protective layer SiO2. At the same time， the relationship between the characteristics of PECVD SiO2 films and the important factors of deposition based on IGZO-TFT was also summarized.

The corner white Mura issue of product “A” produced in line 6 generation was systematically studied. Firstly， the physical analysis was carried out for the defects， and then the data simulation analysis was carried out by using the binary logistic regression method of Minitab software. Finally， the defects were analyzed and improved through different process test conditions. The experimental results showed that the sealant in defect region was sheltered by metal wiring， and the shielding proportion was an important factor affecting the corner white dot. Further analysis by TOF-SIMS showed that defect region had higher content of sealant （acrylic acid， bisphenol A）. Thus， the sealant was sheltered by metal wiring， resulting in incomplete curing of the sealant in this region， leading to the precipitation of the sealant component into the liquid crystal， which then affected the normal deflection of the liquid crystal， resulting in white dot in the corner at last. By adjusting the corner of the sealant painting， it could effectively reduce the defect rate. This experiment effectively improved the corner white dot， laid a firm foundation for the smooth mass production of product “A”， and reduced the loss of the company.

Simulation software was used to simulate the temperature distribution of the module and each component， and the simulation process of heat and the correctness of the simulation results were studied. Firstly， simulation software was used to analyze the heat dissipation of the module to obtain the temperature distribution of the module and each component. Then， the module was placed in a 25 ℃ incubator， and the temperature was measured of the random points on the surface of the module and OC by a multiple temperature detector. Finally， the simulated temperature was compared with the measured temperature， the temperature rise error value was calculated， and the difference between the measured value and the simulation value was compared through the temperature rise error value，to verify the correctness of the simulation results. By comparing and analyzing the experimental results and the simulation results， it was concluded that： when the module operated stably at the room temperature， there was little difference between the simulation value and the actual measured value， and the maximum temperature rise error was 3.5%， which was relatively small. The simulation results had theoretical reference and met the engineering design requirements.

In this paper， aiming at the separation phenomenon of Gamma curve of dual-screen display products， the mechanism was studied through experimental verification such as physical analysis， electrical experiments， process data and adjustment of process conditions， as well as a large number of verification data. Through experimental measurements， it was known that there were differences in the cell gap and the pixel electrode CD between the upper and lower screens， resulting in differences in the transmittance of the upper and lower screens， then resulting in differences in Gamma between the upper and lower screens. Through process investigation and verification， it was found that the difference in the cell gap and the pixel electrode CD was caused by the influence of equipment hardware， resulting in poor uniformity of the adhesive starting position；and finally the process of sealing the upper and lower screens was achieved， the optimized film thickness was introduced and the pixel electrode CD increase process within the fluctuation range was improved. The Gamma curve separation defect was completely improved （defect rate of 12%） and the screen display quality of the product was improved.