Sign in / Sign up
Home > News >
FPGA implementation of ultra-high frame rate image acquisition and transmission system
2024-01-27 20:35:16

LUX1310,LUXIMA Special distributor,1.3MP 1280x1024,1070fps high speed,global shutter 12-bit,2/3-inch CMOS,digital image sensor for high speed machine vision,industrial cameras and biomedical cameras,16 parallel LVDS output,LUX1310 socket set,Socket for LUX1310


LUX1310,LUXIMA Special distributor,1.3MP 1280x1024,1070fps high speed,global shutter 12-bit,2/3-inch CMOS,digital image sensor for high speed machine vision,industrial cameras and biomedical cameras,16 parallel LVDS output,LUX1310 socket set,Socket for LUX1310

With the progress and development of science and technology, high-speed photography plays an increasingly important role in scientific research, industrial manufacturing and daily life.
At present, due to the overall slow development of high-speed cameras in China, and improving the camera frame rate will increase the transmission bandwidth of image data, so improving the camera frame rate has been the bottleneck in the development of high-speed cameras in China. FPGA is the field programmable gate array, it has the advantages of programmable, high integration, high-speed parallel and good stability, especially in the high-speed parallel data transmission advantages, so FPGA is very suitable as the main control chip of high-speed camera. Based on this, this paper chooses to study the FPGA implementation of image acquisition and transmission system of ultra-high frame rate.
Firstly, this paper builds the image acquisition sub-card according to the selected LUX1310 image sensor and FMC connector, uses Cadence software to draw the schematic diagram of the image acquisition sub-card, and finally realizes the image acquisition sub-card. Then the whole scheme is divided into main control logic, dynamic data alignment, non-uniformity correction and optical fiber transmission modules by modularization idea, and the hardware implementation of each module is carried out by Verilog code.
Then, in order to solve the problem of sampling instability caused by delay in high-speed image data output by LUX1310 image sensor, a dynamic data alignment algorithm is proposed, which uses the way of dynamically adjusting the increasing or decreasing delay to detect the edge of data conversion and measure the width of data. The image data is stably sampled at the clock edge to ensure the robust reception of image data. At the same time, aiming at the non-uniformity problem existing in the acquisition of ultra-high frame rate images, a non-uniformity correction algorithm is proposed. The algorithm obtains the bias correction factor by using the single-step cyclic fine-tuning method and the gain correction factor under different illumination intensity, which can effectively improve the non-uniformity problem and improve the image quality.
Finally, an experimental platform with KC705 development board and dual-port 10GbE 10-gigabit optical fiber network card as the main module is built. The register reading and writing, data processing, non-uniformity correction and image acquisition of the ultra-high frame rate image acquisition and transmission system are tested, and the final test shows that the ultra-high frame rate image acquisition and transmission system works properly. The final test achieved a frame rate of 2443fps with a resolution of 640*512 at a bandwidth of 10Gbps. The design can be widely used in the fields of biomedicine, sports, energy and chemical industry, and has a good application prospect.

Prev:The world's first USB-IF PD3.0 certified Canyon CY2311/CY2312 USB Type-C PD charging control chip solution
Next:List of SONY image sensors for industrial cameras, factory automation, intelligent transportation system cameras