A method and apparatus for enabling a user to define a panning and zooming path across a still image in a digital imaging device during interactive movie creation. The method and apparatus include allowing the user to define the position of a plurality of key frames on the still image, allowing the user to size the key frames to control the zoom during the movie, and allowing the user to set the key frames on the still image to fix each of the key frame's position and size.

Method and apparatus for defining a panning and zooming path across a still image during movie creation

This paper presents an advanced video camera system with robust automatic focus (AF), automatic white-balance (AWB), and enhanced zoom tracking. The proposed system can achieve accurate zoom tracking with significantly reduced system memory. It can also find accurate in-focus state even when the camera shoots at a CRT monitor or a light source. The proposed AWB technique compensates the luminance intensity of color components without degrading the image.

A video camera system with enhanced zoom tracking and auto white balance

— Mamdani Fuzzy Model is an important technique in Computational Intelligence (CI) study. This paper presents an implementation of a supervised learning method based on membership function training in the context of Mamdani fuzzy models. Specifically, auto zoom function of a digital camera is modelled using Mamdani technique. The performance of control method is verified through a series of simulation and numerical results are provided as illustrations. Keywords-component: Mamdani fuzzy model, fuzzy logic, auto zoom, digital camera I. I NTRODUCTION The adoption and adaptation of digital signal processing technology in consumer digital cameras has improved the performance of automatic adjustments including auto facus (AF), auto exposure (AE) and zoom tracking, which are important features for high image quality, under various shot conditions [1]. In a digital camera, optical parameters such as zoom, focus and aperture, are usually motor (computer) controlled [2]. The zoom tracking is the continuous adjustment of a camera’s focal length, to keep in-focus state of camera image during zoom operation. Without zoom tracking operation, a camera system cannot maintain in-focus state during zooming. Some conventional zoom tracking techniques were implemented using curve trace stored in a look-up table [3]-[6]. The simple look-up table method, however, requires a large system memory. Moreover, it is difficult to select a proper curve when the zoom lens moves toward the tele-angle. There are several methods that determine the variations of the intrinsic parameters developed in the literature. Recently, some research has been conducted by the employment of neural networks [7]. Nevertheless, this method is too complex to be used and require a lot of training data which might not be always available. The simplest modeling technique is to measure the intrinsic parameters using a standard offline calibration technique at several zoom and focus settings and then store the data in a look-up table [8]. This leads unfortunately to a large memory requirement especially when a precise model is needed. A novel technique using moving least-squares approach to model the variation of the camera internal parameters as a function of focus and zoom was proposed by [9]. Compared to a previous technique using a global least-squares regression scheme with bivariate polynomial functions, the new method resulted in reduction of the mean estimation error. System Modelling based on conventional mathematical tools (e.g., differential equations) is not well suited for dealing with ill-defined and uncertain systems. By contrast, a fuzzy system employing fuzzy if-then rules can model the qualitative aspects of human knowledge and reasoning processes without employing precise quantitative analysis [10]. Increasingly, fuzzy system as a promising Computational Intelligence technique has found many industrial applications. Different fuzzy models have been developed, and successfully applied such as Mamdani model and Sugeno model. There is not a universally best fuzzy model, but each model or may have its suitable types of application. So how to automatically choose an appropriate fuzzy model or for a specific application is always important and sometimes difficult [11]. Interest has been shown recently in the applications in the fields of image recognition, especially for image zooming. Image zooming involves the adjustment of a camera’s focal length to keep in-focus state of camera image during zooming operation. It is well known that there are two types of zooming operations, i.e., optical and digital zoom. Many compact digital cameras can perform both an optical and a digital zoom. A camera performs an optical zoom by moving the zoom lens so that it increases the magnification of light before it even reaches the digital sensor. In contrast, a digital zoom degrades quality by simply interpolating the image- after it has been acquired at the sensor [12]-[14]. Digital Camera involves the zooming function which allows us to focus our target image in variety of distance. Generally, the determination of zoom in (+nX) and zoom out (-nX) is proportional to the distance between the digital camera and the object that want to capture. In this paper, the auto image zooming that was implemented by using Mamdani fuzzy model is described where rules are derived from multiple knowledge sources such as previously published databases and models, existing literature, intuition and solicitation of expert opinion to verify the gathered information. This paper is not concerned with theoretical discussion of digital camera zooming functions, rather, it concentrates on the application Mamdani fuzzy model in the zooming functions of digital camera. Preliminary results of this study were published in CIM conference proceedings [15].

The Application of Mamdani Fuzzy Model for Auto Zoom Function of a Digital Camera

Zoom tracking is becoming an integral part of any digital still camera. Zoom tracking involves the automatic adjustment of the focus motor in response to the zoom motor movements for the purpose of keeping an object of interest in focus. In this paper, we have introduced a new zoom tracking approach, named relational zoom tracking, where an estimate of a trace curve is obtained by using so called relational curves. A trace curve denotes focus motor positions versus zoom motor positions for a given object distance, whereas a relational curve denotes focus motor positions versus object distances for a given zoom motor position. The hardware implementation of this algorithm on a digital still camera is also provided in this paper. The obtained results show that the developed zoom tracking approach provides higher tracking accuracies as compared to the conventional zoom tracking methods while generating comparable or better tracking speeds.

A relational approach to zoom tracking for digital still cameras

To achieve size preserving tracking, in addition to controlling the camera's pan and tilt motions to keep the object of interest in the camera's field of view (FOV), the camera's focal length is adjusted automatically to compensate for the changes in the target's image size caused by the relative motion between the camera and the target. The estimation accuracy of these changes determines the effectiveness of the resulting zoom control. The existing method of choice for real-time target scale estimation applies structure from motion (SFM) based on the weak perspective projection model. In this paper we propose a target scale estimation algorithm with a linear solution based on the more advanced paraperspective projection model, which improves the accuracy of scale estimation by considering center offset. Another key issue in SFM based algorithms is the separation of target and background features, especially when composite camera (pan/tilt/zoom) and target motions are involved. This paper designs a fast target feature separation/grouping algorithm, the 3D affine shape method. The resulting separation automatically adapts to the target's 3D geometry and motion and is able to accommodate a large amount of off-plane rotation, which most existing separation/grouping algorithms find difficult to achieve. Experimental results illustrate the effectiveness of the proposed scale estimation and feature separation algorithms in tracking translating and rotating objects with a PTZ camera while preserving their sizes. In comparison with the leading size preserving tracking algorithm described by Tordoff and Murray, our algorithm is able to reduce the cumulative tracking error significantly from 17.4% to 3.3%.

3D Target Scale Estimation and Target Feature Separation for Size Preserving Tracking in PTZ Video

In this paper we consider the use of the GETAFIX robot vision head for object recognition by automatic active panning, zooming and focussing. It is shown that given appropriate areas of interest (which may be defined by motion, change detection or colour segmentation) the sensor can be driven to isolate these regions. Subsequent processing is made more reliable in this process. The effective signal to noise ratio of the data is increased by the process and many features which were unreliably detected by passive techniques can be reliably and accurately acquired by the active approach. Experiments are demonstrated for the case of detecting cylindrical 3D objects in table-top scenes.

Automatic control of camera pan, zoom and focus for improving object recognition

Recognition of 3D Cylinders in 2D Images by Top-Down Model Imposition

A computer system identifies a dispatching rule that corresponds to a dispatching decision to be made in a simulation of a production environment. The dispatching rule is associated with a set of data processes. The computer system identifies a subset of the data processes that is to be executed in the simulation and executes the subset of data processes to simulate the dispatching rule to make the dispatching decision. The computing system generates a simulation decision that describes the dispatching decision made from executing the subset of data processes.

Configuring a dispatching rule for execution in a simulation

Object recognition has been one of the main areas of research into computer vision in the last 20-30 years. Until recently most of this research has been performed on scenes taken using static monocular, binocular or even trinocular cameras. It is believed, however, that by adding the ability to move the look point and concentrate on a region of interest a more robust and efficient method of vision can be achieved. Recent studies into the ability to provide human-like vision systems for a more active approach to vision have lead to the development of a number of robot controlled vision systems. In this thesis the development of one such system at the University of Surrey, the stereo robot head "Getafix" is described. The design, construction and development of the head and its control system have been undertaken as part of this project with the aim of improving current vision tasks, in particular, that of object recognition. In this thesis the design of the control systems, kinematics and control software of the stereo robot head will be discussed. A number of simple commissioning experiments are also shown, using the concepts of the robot control developed herein. Camera lens control and calibration is also described. A review of classical primitive based object recognition systems is given and the development of a novel generic cylindrical object recognition strategy is shown. The use of this knowledge source is demonstrated with other vision processes of colour and stereo. The work on the cylinder recognition strategy and the stereo robot head are finally combined within an active vision framework. A purposive active vision strategy is used to detect cylindrical structures, that would otherwise be undetectable by the cylindrical object detection algorithm alone.

Paul. Hoad

Papers

Automatic control of camera pan, zoom and focus for improving object recognition

Recognition of 3D Cylinders in 2D Images by Top-Down Model Imposition

Configuring a dispatching rule for execution in a simulation

Active robot vision and its use in object recognition.