Industrial Technology Research Institute

National Chung Cheng University

Oscal T-c Chen

Ruey-Liang Ma

Jeng-Yih Wang

Wei-Jean Liu

Kuo-Hua Chen

Hsiu-Fen Yeh

The present invention discloses an apparatus of high dynamic-range CMOS image sensor and method thereof. The present invention utilize a pixel circuit outputting an output signal, wherein the pixel circuit has a photodiode and a plurality of transistors; and utilize a current source as a charge supplement unit to supply current into one end of the photodiode, and providing charges to the parasitic capacitor of the photodiode to delay saturation in the pixel circuit. In addition, a feedback circuit can be further designed connecting the pixel circuit. The feedback circuit receives the output signal from the pixel circuit and then outputs a control signal according to the output signal of the pixel circuit to control status of the charge supplement unit, and thereby increasing the dynamic range of the CMOS image sensor.

Apparatus of high dynamic-range CMOS image sensor and method thereof

An energy saving multiplication device and its method is disclosed. The multiplication device includes a dynamic range determination unit, a Booth encoding/decoding unit and a counter array. The dynamic range determination unit determines dynamic ranges of the numerical values to be multiplied together and outputs after processing according to the dynamic-range size relation of the input data. The Booth encoding/decoding unit couples to the dynamic range determination unit. The counter array couples to the Booth encoding/decoding unit for accumulating the partial products to obtain the products of the input data.

Energy saving multiplication device and method

A programmable finite impulse response processor, by which a convolution calculation between input data and filter coefficients is performed based on Booth algorithm. The processor include a pre-processing unit, data latches, a configurable connection unit, Booth decoders, coefficient registers, a path control unit, and a post-processing unit. The pre-processing unit is used to partition the input data into a pipeline sequence which include a plurality of sequence units in a Booth format. According to both the dynamic ranges of the input data and of the filter coefficients, the configurable connection unit is used to select certain parts of the sequence units for the convolution calculation, so that a dynamic data range is scaleable by the processor. By Booth decoders, the selected sequence units are decoded and multiplied by corresponding filter coefficients stored in the coefficient registers. By the path control unit, the bit lengths of the filter coefficients are scaled and accumulation pathes are selected. The post-processing unit is to perform a final accumulation of the convolution calculation results between the selected sequence units and the filter coefficients.

Growing community of inventors

Taipei, Taiwan

Oscal T-c Chen