Photographic camera simulation systems working from computer memory

• US 4,731,864 A
• Filed: 02/21/1986
• Issued: 03/15/1988
• Est. Priority Date: 02/21/1986
• Status: Expired due to Term

• Current Assignee: Intel Corporation, RCA Corporation
• Original Assignee: RCA Corporation
• Inventors: Modla, Andrew A.
• Primary Examiner(s): NOT, DEFINED
• Assistant Examiner(s): Brim, Steven W.
• Application Number:
  US06/831,561
• Time in Patent Office: 753 Days
• Field of Search: 358/103, 358/104, 358/109, 358/125, 364/522, 364/518, 364/521, 382/41, 382/54
• US Class Current: 382/255
• CPC Class Codes: G06T 15/00 3D [Three Dimensional] image rendering
• CPC Class Codes: G09B 9/302 the image being transformed by computer processing, e.g. updating the image to correspond to the changing point of view

Abstract

A camera field-of-view is mapped with full depth-of-focus into electronic memory. Data concerning image intensities at selected points in image space are stored, along with data concerning the distances between the camera and these points in image space. Images with any focus and depth-of-focus can be generated by scanning the electronic memory during a reading thereof and convolving the data concerning image intensity with an adjustable-bandwidth low-pass spatial filter. The bandwidth of that filter is adjusted dependent on the data concerning distances between camera and points in image space. This image storage and image retrieval procedure conserves the size of electronic memory required to store and retrieve images with any focus and depth-of-focus.

Claims

1. Apparatus for electronically generating reduced depth-of-focus images, comprising:

a memory having storage locations addressed in two-dimensions which map two orthogonal dimensions in an image space, each addressable storage location in said memory storing a digital word, a first portion of which word is a sample of the intensity of full-depth-of-focus image, and a second portion of which word is the distance between the camera and the portion of said full-depth-of-focus image described by the first portion of that word;

means for reading from storage locations in said memory in a sequence corresponding to a scanning of image space;

means convolving the first word portions read from storage locations in said memory with an adjustable-bandwidth low-pass spatial filtering kernel for generating samples of a selected one of said reduced-depth-of-focus images in which foreground and background pixels are blurred and their blurs are superposed on surrounding pixels; and

means for adjusting, in accordance with a selected one of prescribed relationships, the spatial bandwidth of said low-pass spatial filtering kernel responsive to the second word portions read from storage locations in said memory.

2. Apparatus as set forth in claim 1 wherein said means for adjusting the spatial bandwidth of said low-pass spatial filtering kernel includes:

means for calculating from the second word portions read from storage locations in said memory how the blur of each background pixel is eclipsed by other pixels; and

means for adjusting the spatial bandwidth of said low-pass filtering kernel to provide for eclipsing the blur of background pixels by said other pixels.

3. A method for electronically generating images of controlled focus and depth-of-focus, said method comprising the steps of:

mapping a camera field-of-view in memory, providing for each of selected points in image space a datum concerning the intensity of image at that point and a datum concerning the distance between the principal axis of the camera lens and the object plane of that point;

scanning said memory during reading thereof;

convolving an adjustable-bandwidth low-pass spatial filtering kernel and the data that scanning the memory during reading thereof supplies concerning the intensity of image, for generating samples of said images of controlled focus and depth-of-focus; and

adjusting the spatial bandwidth of said low-pass spatial filtering kernel responsive to the data that scanning the memory during reading thereof supplies concerning the distance between the principal axis of the camera lens and the object plane.

4. A method for electronically generating images as set forth in claim 3 including the further steps of

comparing, the distance between the principal axis of the camera lens and the object plane of that sample in image space at which the center of said low-pass spatial filtering kernel currently falls during said step of convolving, with the distances between the principal axis of the camera lens and the object planes of surrounding samples;

generating an eclipsing mask based on those comparisons;

ascertaining when a datum concerning the intensity of image is associated with a datum concerning the distance between the camera lens and object plane greater than that associated with good focus, thereby to identify a background pixel; and

modifying said low-pass spatial filtering kernel by said eclipsing mask whenever a background pixel is so identified.