Radiant energy – Photocells; circuits and apparatus – Photocell controlled circuit
Reexamination Certificate
2000-01-19
2003-05-13
Pyo, Kevin (Department: 2878)
Radiant energy
Photocells; circuits and apparatus
Photocell controlled circuit
C348S275000
Reexamination Certificate
active
06563101
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates generally to image tracking devices. More specifically, the present invention relates to image tracking devices including an array of light-sensitive elements disposed in a non-rectilinear geometry for tracking an image incident on the array.
Articles and publications set forth herein are presented for the information contained therein: none of the information is admitted to be statutory “prior art” and we reserve the right to establish prior inventorship with respect to any such information.
Image sensor arrays are commonplace in modern electronic devices such as displacement sensors, digital cameras, copiers, scanners, facsimile machines, and camcorders, for example. An image sensor array can be a one-dimensional array having a row of light-sensitive elements, or the array can be a two-dimensional array in which there are a plurality of light-sensitive elements disposed in rows and columns of the array. In either case, the array is laid out in a Cartesian geometry (rectilinear geometry) with the light-sensitive elements arranged in an orderly and regular pattern throughout the array. Moreover, the light-sensitive elements are structurally identical to one another and are merely replicated throughout the array. Resulting is a fill-factor that is constant throughout the array. A fill-factor is the ratio of the active light sensing area of a light-sensitive element to the full physical area of the array element.
Image sensor arrays in modern imaging devices such as digital cameras and camcorders, for example, are laid out in a rectilinear geometry. The rectilinear geometry is dictated by semiconductor layout design rules that traditionally require orthogonally arranged circuit elements in order to facilitate semiconductor fabrication. A charge-coupled device (CCD) and a CMOS active pixel sensor are exemplary image sensor arrays that are laid out in a rectilinear geometry.
An optical system that includes one or more lenses can be used to focus an image onto the sensor array. The image can have geometrical distortions such as pincushion distortion, or barrel distortion that are produced by the optical system or by curvature of the object surface being imaged. Additionally, the image can have other field distortions including non-uniform field illumination and non-uniform image resolution. Reducing distortions traditionally requires a lens designer to incorporate additional lens elements, complex lens surface shapes (aspheric lenses), apertures, different or additional glass types, or even filters.
Because the rectilinear geometry of the sensor array has been an accepted design constraint, the burden of correcting the distortions has been placed largely on the optical system. Therefore, as a result, the design variables (the degrees of design freedom) available to the lens designer have been limited to those that can be addressed by changes to the optical system only. However, correcting optical distortions has come at a price, namely, increased cost, weight, size, and complexity of the optical system. Generally, it is desirable to keep the cost of digital imaging devices as low as possible, particularly for products aimed at a consumer mass market. Furthermore, some digital imaging devices are designed to be portable, adding additional design constraints of low weight and small size.
Consequently, there is a need to increase the number of design variables and thereby the degrees of design freedom available to the optics and sensor systems designer. Because geometric distortions are non-rectilinear, one way to increase the number of design variables or degrees of design freedom in the lens and sensor system is to eliminate the design constraint of rectilinear geometry in the sensor. A sensor array having a non-rectilinear geometry can be used to correct geometric distortions in images, reducing the design requirements of the lens. Furthermore, the fill-factors of the light-sensitive elements can be varied with field position to address non-uniform field illumination. And array cell size can be varied with field position to match non-uniform image resolution.
A clever use for image sensor arrays is within sensing devices that track motion of an object. A pattern on a surface of an object is imaged onto an image sensor so that motion of the object causes an image of the surface pattern to traverse an image sensing array.
In a rotary encoder, for example, surface or transmission features of an encoder disc manifest patterns in an image at a detector. Rotation of the encoder disc causes the patterns in the image to traverse an image sensor that is disposed in the rotary encoder. For instance, a light source, such as a light-emitting diode (LED), is used to illuminate the encoder disc so that light from the LED is reflected, scattered off of, or transmitted through the encoder disc to form the image that is incident on the image sensor. The image sensor is laid out in a one-dimensional rectilinear geometry. For example, two to four elements comprising the image sensor can be arranged in a one-dimensional layout (in a straight line).
An example of this in the present art is the use of quadrature signals such as used in shaft encoders to sense angular displacement and direction of a rotating shaft that is coupled to the encoder. However, one disadvantage to a one-dimensional layout of the image sensor when using more than two elements is that coordinate transformation algorithms are required to convert data from the image sensor into data representative of the two-dimensional motion of the image, if maximum resolution is desired.
Another disadvantage of the systematic regularity of a conventional rectilinear layout is that high spatial frequencies in the image can get aliased into the lower spatial frequencies of the data from the image sensor if the distance or pitch length between the elements is more than half the period of those higher frequencies in the image. Aliasing is a disadvantage because it leads to incorrect tracking results.
Therefore, there is a need to overcome disadvantages associated with rectilinear image sensors. Non-rectilinear arrays can be laid out with curvilinear rows and/or columns. For example, when trajectories of object motion are constant as in simple rigid-body motion or in laminar flow, either the rows or columns can parallel the trajectories of image features that move across the array, even in the presence of image distortions. Such a non-rectilinear image sensor array can be used to track a certain two-dimensional motion of an image (incident on the array) without the need to apply coordinate transformation algorithms to make corrections in the post-processing of image data from the image sensor. Furthermore, the spacings between adjacent elements comprising an image sensor array, the position of each active photo sensor element within its cell of the array, and the fill-factor of each element of the array can be adjusted by design to frustrate aliasing.
SUMMARY OF THE INVENTION
The problems and limitations associated with rectilinear image sensors are addressed by various aspects of the present invention. The problem with geometrical distortions inherent in an optical system is addressed by positioning photosensors in an array having a non-rectilinear geometry. The array can have a shape that compensates for anticipated geometric distortion in an image; for example, the array can have a pincushion geometry to compensate for pincushion distortion in the image. Problems with field-dependent illumination and resolution are addressed by varying the sizes of the array elements and/or by varying the fill-factors of individual elements. Fill-factor is defined as the ratio of light-sensitive area within an array element to the area of the whole array element. For example, in regions of the image where illumination fall-off is present, the active areas of the light-sensitive elements in those regions can be increased to compensate for the illumination fall-off (relative to active area sizes in regions w
Denny III Trueman H.
Pyo Kevin
Sohn Seung C.
LandOfFree
Non-rectilinear sensor arrays for tracking an image does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Non-rectilinear sensor arrays for tracking an image, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Non-rectilinear sensor arrays for tracking an image will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3030728