A large variety of solutions for widening the dynamic range (DR) of CMOS image sensors has been proposed throughout the years. We propose a set of criteria upon which an effective comparative analysis of the performance of wide-DR (WDR) sensors can be done. Sensors for WDR are divided into seven categories: 1) companding sensors; 2) multimode sensors; 3) clipping sensors; 4) frequency-based sensors; 5) time-to-saturation (time-to-first spike) sensors; 6) global-control-over-the-integration-time sensors; and 7) autonomous-control-over-the-integration-time sensors. The comparative analysis for each category is based upon the quantitative assessments of the following parameters: signal-to-noise ratio, DR extension, noise floor, minimal transistor count, and sensitivity. These parameters are assessed using consistent assumptions and definitions, which are common to all WDR sensor categories. The advantages and disadvantages of each category in the sense of power consumption and data rate are discussed qualitatively. The influence of technology advancements on the proposed set of criteria is discussed as well.

Wide-Dynamic-Range CMOS Image Sensors—Comparative Performance Analysis

A CMOS image sensor that can operate in both linear and logarithmic mode is described. Two sets of data are acquired and combined in the readout path to render a high dynamic range image. This is accomplished in real-time without the use of frame memory. A dynamic range in excess of 120 dB was achieved at 26 frames/s (352times288-array). The system addresses the problems of high fixed pattern noise (FPN), slow response time, and low signal-to-noise ratio (SNR) in logarithmic mode. FPN has been effectively reduced by single and two parameter calibration, the latter achieving FPN of 2% per decade. A novel on-chip method of deriving a reference point has been implemented. The system is fabricated in a 0.18-mum 1P4M process and achieves a pixel pitch of 5.6 mum with 7 transistors per pixel

/pdf/extended-dynamic-range-from-a-combined-linear-logarithmic-5cboy0yt9m.pdf

Extended Dynamic Range From a Combined Linear-Logarithmic CMOS Image Sensor

A combined linear and logarithmic image sensor is implemented in a 0.35 /spl mu/m 1P3M technology. The pixel is 7.5/spl times/7.5 /spl mu/m/sup 2/ with a 37% fill factor and contains only 4 transistors. Offset calibration in the logarithmic region is realized by using electrical charge injection into the photodiode. The sensor achieves 120 dB DR and the offset calibration reduces the FPN from 13 mV to 5 mV.

A linear-logarithmic CMOS sensor with offset calibration using an injected charge signal

A high dynamic range CMOS image sensor providing a user-programmable power responsivity curve is presented. Each pixel integrates, besides a 4T active pixel structure, a voltage comparator and an analog memory to implement a time-to-saturation scheme while also providing the standard integrated photo-current signal. The sensor generates two 10-bit analog outputs allowing a typical dynamic range exceeding 120 dB with a temporal noise lower than 0.13% and a fixed pattern noise of 0.4% (1.7%) of the total signal swing (2 V) at low (high) irradiance without any external calibration procedures. A 140 times 140-pixel array has been fabricated in a 0.35-mum, two-poly four-metal (2P4M), 3.3-V standard CMOS technology. The chip measures 3.9 times 4.6 mm2 with a pixel pitch of 15 mum and a fill factor of 20%.

A 120-dB Dynamic Range CMOS Image Sensor With Programmable Power Responsivity

All things considered, electronic imaging systems do not ri- val the human visual system despite notable progress over 40 years since the invention of the CCD. This work presents a method that allows design engineers to evaluate the performance gap between a digital camera and the human eye. The method identifies limit- ing factors of the electronic systems by benchmarking against the human system. It considers power consumption, visual field, spatial resolution, temporal resolution, and properties related to signal and noise power. A figure of merit is defined as the performance gap of the weakest parameter. Experimental work done with observers and cadavers is reviewed to assess the parameters of the human eye, and assessment techniques are also covered for digital cam- eras. The method is applied to 24 modern image sensors of various types, where an ideal lens is assumed to complete a digital camera. Results indicate that dynamic range and dark limit are the most lim- iting factors. The substantial functional gap, from 1.6 to 4.5 orders of magnitude, between the human eye and digital cameras may arise from architectural differences between the human retina, arranged in a multiple-layer structure, and image sensors, mostly fabricated in planar technologies. Functionality of image sensors may be signifi- cantly improved by exploiting technologies that allow vertical stacking of active tiers. © 2011 SPIE and IS&T. (DOI: 10.1117/1.3611015)

/pdf/toward-a-digital-camera-to-rival-the-human-eye-amllrmuj6m.pdf

Toward a digital camera to rival the human eye

A 352/spl times/288 pixel array achieves >120 dB dynamic range through merging sequential linear and logarithmic images Calibration is used to match offset and gain A 7-transistor pixel is built in a 018 /spl mu/m 1P4M CMOS process

G. G. Storm

Papers

Extended Dynamic Range From a Combined Linear-Logarithmic CMOS Image Sensor

Combined linear-logarithmic CMOS image sensor

High dynamic range imaging using combined linear-logarithmic responses from a CMOS image sensor