A non-linear column-level ADC and method for improving the contrast of cmos image sensors

Abstract

The invention discloses a nonlinear column-level ADC for improving the contrast of a CMOS image sensor, which includes a nonlinear ramp generator connected to a counter as the same clock input, the output end of the ramp generator connected to the positive input end of a comparator, and the negative input end of the comparator The terminal is connected to the pixel voltage sampling, the output terminal of the comparator is connected to the latch and the register in turn, and the register is also connected to the counter, and the register is used to save the value of the counter and output the result. The invention also discloses a method for improving the contrast of the CMOS image sensor The method not only improves the contrast of the CMOS image sensor, but also can effectively identify the cell image under the lensless imaging system.

Description

technical field [0001] The invention belongs to the technical field of high-contrast CMOS image sensors, and in particular relates to a nonlinear column-level ADC for improving the contrast of CMOS image sensors. [0002] The invention also relates to a method for improving the contrast of the CMOS image sensor. Background technique [0003] With the development of the field of biomedical diagnostic testing, cell detection technology provides an important basis for diagnosing the life status and basic information of organisms. Different physical conditions will correspond to different cell types, cell numbers, cell structures, and cell shapes. Understanding the various characteristics of these cells is of great significance for human health, disease research, and drug development. The most basic cell detection methods at this stage include chemical detection and optical microscope detection, but their operations are complicated and require professionals to operate, which ca...

AI Technical Summary

Problems solved by technology

[0004] In the lensless cell imaging system, although the performance of the CMOS image sensor used has been continuously improved with the development of integrated circuits, most of them have a linear and uniform output, which is suitable for environments with bright and uniform lighting, but in dark light without lenses When imaging cells in an environment, due to their low sensitivity to light intensity, darker images will be obtained

In addition, under the lensless cell imaging system, the local contrast of the obtained cell image is also low, because the transmittance of the main cell is close to the transmittance of the surrounding liquid background, so the gray distribution range of the image is narrow and slightly blurred, and the local area is gray. The contrast values ​​are relatively close, resulting in a decrease in contrast and making it difficult to quickly distinguish image content

To solve this problem, three solutions are usually adopted at this stage: First, add an image signal processing circuit to the ADC to expand the dynamic range by introducing an automatic exposure logic circuit. However, when this method is used in a bright light environment, the conversion will be saturated and the dynamic The range becomes smaller, and the added ISP circuit will also occupy the area of ​​the chip and increase power consumption

Second, multi-frame exposure or multiple related sampling methods are used, but multiple sampling will affect the frame rate of the system; and multi-frame exposure will increase the exposure time, resulting in prolonged sensor shooting time

Third, the digital contrast stretching algorithm is used outside the chip. Although this method is easy to implement, it has two disadvantages. The first is that it is mainly aimed at the overall image. The higher the requirements, the more complex the mapping function, which is difficult to implement in CMOS images. The integration inside the sensor; the second is that digital correction will amplify the quantization error under low illumination, and the image details in the illumination range will be affected at this time

These methods increase the complexity of circuit design to varying degrees, increase circuit consumption, and are difficult to apply to the needs of future CMOS image sensors.

Images