X-Google-Language: ENGLISH,ASCII-7-bit X-Google-Thread: f996b,4efec4018e878d11,start X-Google-Attributes: gidf996b,public X-Google-ArrivalTime: 2002-09-09 06:32:59 PST Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news-spur1.maxwell.syr.edu!news.maxwell.syr.edu!newsfeed1.cidera.com!Cidera!nntp.magma.ca!news!not-for-mail Message-ID: <3D7CA253.6AB6A0E7@magma.ca> From: Big Picture Effects X-Mailer: Mozilla 4.79 [en] (Win98; U) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: alt.ascii-art Subject: Image to ASCII algorithms Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Lines: 64 Date: Mon, 09 Sep 2002 13:30:53 GMT NNTP-Posting-Host: 64.26.147.162 X-Trace: news 1031578253 64.26.147.162 (Mon, 09 Sep 2002 09:30:53 EDT) NNTP-Posting-Date: Mon, 09 Sep 2002 09:30:53 EDT Organization: Magma Communications Ltd. Xref: archiver1.google.com alt.ascii-art:19429 Dear readers, I am researching Image to ASCII conversion programmes, and I need some help on the algorithms. My intention is not to write a new image to ASCII conversion programme, rather, I would like to learn about the background and about types of algorithms and how long they have been around (70's, 80's 90's?). My background is image processing. Could someone shed some light on which algorithm is used in which programme? Could someone send me excerpts of the source? Please contact me directly if you are a software author, and if you are uncomfortable talking about the precise tricks or sending source code. I am sure we can work something out to warrant your copyright, authorship and of course artistic originality. Perhaps you could refer me to the experts, or point to algorithm descriptions and specific source code? If I know where to find specific source code, I would be able to hunt down the programme with more vigor! Email me directly at bigpicture@magma.ca. I will be happy to summarize the findings and post it back to the news group. For starts, I am unfamiliar with the programme internals, but I could think of a few algorithms. An image is split into M-by-N tiles (sub images) and then 1. the gray values for each tile is computed, and an M-by-N array of ASCII characters is generated, whereby for each tile the ASCII character that best matches the grey-value is placed in the array; pre-calculated grey-values of selected ASCII characters are used for the search, while the actual ASCII character is used in the final composition; or 2. as above, but now the bit map of the ASCII characters are used, not just an overall grey-value; the tile size (pixels) is made to match the the bit map size (pixels) by image resizing (e.g. pixel averaging or interpolation) 3. as above, but now more refined images of characters are used, not just a binary bit map, and the images contain alternate backgrounds, colors and perhaps even different font types or font drawings; the images ar used for the search as well as for the final composition. I am posting this request since I know that there are some great non-PC based programmes, and I have no access to Unix or Linux to try them out. Also, I have had some bad luck with web sites that offer downloads of conversion programmes. The description and, if possible, the source code of the core algorithm is really all I would need to continue with my research. Your input would be very valuable, Thanks, Aryan.