Mark Gamadia

A new auto-focus sharpness function for digital and smart-phone cameras

Passive auto-focusing is a key feature in consumer level digital and smart-phone cameras and is used to capture focused images without any user intervention. This paper introduces a new sharpness function for achieving passive auto-focusing, where the image sharpness information is used to bring it into focus. A comparison is made between this introduced sharpness function and the commonly used sharpness functions in terms of accuracy and computation time. The results obtained indicate that the introduced sharpness function possesses a comparable accuracy while demanding less computation time.

Low-Light Auto-Focus Enhancement for Digital and Cell-Phone Camera Image Pipelines

Images captured by a digital or cell-phone camera in low-light environments usually suffer from a lack of sharpness due to the failure of the cameras passive auto-focus (AF) system to locate the peak in-focus position of a sharpness function that is extracted from the image. In low-light, the sharpness function becomes flat, making it quite difficult to locate the peak.In this paper, a systematic approach is introduced to address the problem of low-light AF by performing computationally simple image enhancement preprocessing steps as part of the image pipeline. These enhancement steps elevate the sharpness function peak, leading to auto-focusing in low-light conditions. A sharpness junction quality measure along with experimental guidelines are presented for determining the most prominent enhancement steps for low-light AF. The implementation results on an actual digital camera platform are also shown to demonstrate the effectiveness of our solution.

Real-time implementation of autofocus on the TI DSC processor

This paper discusses the real-time implementation of a fast and accurate auto-focus method on the Texas Instruments DM270, a programmable processor designed specifically for digital still cameras. The DM270s programmable auto-focus hardware filter is utilized to obtain a sharpness function from a captured image. This function is then used to drive a rule-based search algorithm, which varies the focusing step size depending on the slope of the sharpness function. This leads to faster focusing speeds as compared to the standard global search algorithm. A wide variety of filters are tested by examining their performances in terms of focusing accuracy. The results show that the filters approximating the first derivative operator generate the best focusing accuracy under various focusing conditions.

A filter-switching auto-focus framework for consumer camera imaging systems

Auto-focus (AF) has become a standard feature in consumer level cameras. The most common AF approach analyzes a captured image with a digital band-pass filter applied to a focus region producing a focus value, which is then used by a search algorithm to locate the peak value position. The existing literature does not adequately address how to set the AF parameters such as the filter pass-band and the step-size magnitude. In this paper, a framework is presented to derive the AF parameters automatically based on camera specifications. Also, a new AF search algorithm named Filter-Switching Search is introduced which takes advantage of the derived AF parameters. The effectiveness is demonstrated via real-time implementation on three different camera platforms. Results indicate that the developed solution is portable across different platforms and outperforms the existing approaches without the need for extensive parameter tuning.

Real-time Face-based Auto-Focus for Digital Still and Cell-Phone Cameras

Auto-focus (AF) is a common feature in consumer level digital and cell-phone cameras. Face-based AF, or AF based on face detection, has become of interest due to the fact that the majority of pictures captured by consumers are of human faces. While many face detection algorithms exist in the literature, very few of them are actually suitable for real-time deployment on resource limited digital or cell-phone camera processors. In this paper, a face-detection algorithm combining a Gaussian skin color model with a computationally efficient sub-block postprocessing scheme is introduced to address the realtime constraints encountered in digital and cellphone cameras. This approach has been implemented in conjunction with our previously developed rule- based AF method in order to achieve real-time faced- based AF on the Texas Instruments programmable TMS320DM350 digital camera processor.

Real-time implementation of single-shot passive auto focus on DM350 digital camera processor

With the introduction of high mega-pixel image sensors and large focal length lenses in todays consumer level digital still cameras, single-shot passive auto-focus (AF) performance in terms of speed and accuracy remains to be a critical issue among camera manufacturers. To address the AF performance issue, this paper covers the real-time implementation of a previously developed modified rule-based single-shot AF search method on the Texas Instruments TMS320DM350 processor. It is shown that a balance between AF speed and accuracy is needed to meet the real-time constraint of the digital camera system. Performance results indicate that this solution outperforms the standard global search method in terms of AF speed and accuracy.

Enhanced low-light auto-focus system model in digital still and cell-phone cameras

In low-light conditions, the focus value curve or sharpness function that is used in passive auto-focusing becomes flat due to the increase in the noise level. This paper presents an enhanced low-light auto-focusing system model which provides performance improvements over our previously introduced low-light autofocusing model. This is achieved by utilizing an indexed family of focus value curves via a digital band-pass filterbank. A new sharpness ratio metric is presented for evaluation of auto-focusing in low-light conditions. Experimental results obtained from three different prototype camera platforms demonstrate the improved low-light performance over our previous model.

Real-time auto white balancing using DWT-based multi-scale clustering

Auto white balancing (AWB) involves the process of making white colors to appear as white under different illuminants in digital imaging products such as digital still cameras. This paper presents a computationally efficient auto white balancing algorithm for real-time deployment in imaging products. The algorithm utilizes DWT (discrete wavelet transform) to perform multi-scale clustering (MSC), thus generating a computationally efficient implementation of the original MSC algorithm. The paper also discusses the steps taken to allow running this algorithm in real-time on a digital camera processor. The results of an actual implementation on the Texas Instruments TMS320DM320 processor are provided to illustrate the effectiveness of this algorithm in identifying an illuminant as compared to the widely used gray-world auto white balancing algorithm.

Image Restoration Preprocessing for Low Light Auto-Focusing in Digital Cameras

The performance of passive auto-focus (AF) systems suffers when operating under low-lighting conditions due to the flatness of the sharpness function. In this paper, an adaptive noise reduction method prior to AF sharpness filter is introduced for the purpose of elevating the suppressed peak in the sharpness function . Experimental results show that the use of adaptive noise reduction preprocessing enables focusing at lower lux levels as compared to non-adaptive noise reduction preprocessing.

Performance metrics for auto-focus in digital and cell-phone cameras

Passive auto-focus (AF) is a key component of many digital and cell-phone camera systems. In order to make informed choices in the selection of a passive AF search algorithm, it is essential to have performance metrics. The performance metrics currently utilized do not take into consideration all aspects of auto-focusing. In this paper, four performance metrics which measure all relevant aspects of an AF system is considered at the same time. These metrics include focusing speed, accuracy, power consumption and user experience. Experimental results on a prototype digital camera platform are presented to compare the AF performance of four popular AF search algorithms exhibiting the usefulness of the introduced set of AF performance metrics.