Video compression encoder
View Patent ↗A video compression system is disclosed that is optimized to take advantage of the types of redundancies typically occurring on computer screens and the types of video loss acceptable to real time interactive computer users. It automatically adapts to a wide variety of changing network bandwidth conditions and can accommodate any video resolution and an unlimited number of colors. The disclosed video compression encoder can be implemented with either hardware or software and it compresses the source video into a series of data packets that are a fixed length of 8 bits or more. Sequences of one or more of these packets create unique encoding “commands” that can be sent over any network and easily decoded (decompressed) with either software or hardware. The commands include 3 dimensional copying (horizontal, vertical and time) and unique efficiencies for screen segments that are comprised of only two colors (such as text). Embodiments are also disclosed that improve the video compression depending on the popularity of pixel colors.
1. A method in a data processing system of compressing video having a color depth greater than two colors, comprising:
comparing some portions of selected video frames having a color depth greater than two colors to other portions of the selected video frames;
encoding the some portions of the selected video frames based on the some portions' locational relationship within the selected video frames relative to the other portions;
identifying additional portions of the same selected video frames as pixels comprised only of two colors from a two-color set; and
encoding the additional portions of the selected video frames as a series of bits having binary states corresponding to the two colors in the two-color set.
2. The method of claim 1 , wherein the locational relationship is one from the group consisting of:
(a) a left location within a common frame;
(b) an above location within a common frame; and
(c) a common location within a prior frame.
3. The method of claim 1 , wherein the left location within a common frame is a left location one immediate pixel to the left of a current pixel.
4. The method of claim 1 , wherein the above location within a common frame is an above location one immediate pixel above a current pixel.
5. The method of claim 1 , wherein the video having a color depth greater than two colors is one from the group consisting of:
(a) 24-bit color;
(b) 15-bit color;
(c) 7-bit gray-favored color; and
(d) 5-bit gray-favored color.
6. The method of claim 1 , wherein the locational relationship is one from the group consisting of:
(a) a left location within a common frame;
(b) an above location within a common frame; and
(c) a common location within a prior frame; and
the video having a color depth greater than two colors is one from the group consisting of:
(a) 24-bit color;
(b) 15-bit color;
(c) 7-bit gray-favored color; and
(d) 5-bit gray-favored color.
7. The method of claim 1 , further including:
reading a serial string of pixel color data values for the video and recording information corresponding to the pixel color data values for at least one video frame of the video;
wherein the step of comparing some portions of the selected video frames to other portions of the selected video frames includes comparing current pixel color data values in the serial string of pixel color data values with at least some of the recorded information corresponding to the pixel color data values;
and wherein the encoding of the some portions of the selected video frames based on the some portions' locational relationship within the selected video frames relative to the other portions includes encoding based on the comparison of current pixel color data values with the recorded information.
8. The method of claim 7 , wherein the identification of additional portions of the same selected video frames as pixels comprised only of colors from a two-color set is based on the reading of the serial string of pixel color data values, and the encoding the additional portions as a series of bits having binary states corresponding to the two colors in the two-color set is an encoding based on (1) one binary state corresponding to one pixel color data value read from the serial string and (2) another binary state corresponding to a different pixel color data value read from the serial string.