17-11-2025, 05:28 PM
17-11-2025, 05:28 PM
17-11-2025, 05:37 PM
(17-11-2025, 08:40 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.Here is the description of f101r: You are not allowed to view links. Register or Login to view.
It mentions three cases of 'some similarity'.
Thanks! But, unless I again got the wrong page numbers, I don't see those pairs with You are not allowed to view links. Register or Login to view. as definitely "the same drawing".
On f102r1: You are not allowed to view links. Register or Login to view. there are two cases that 'appear to be the same plant' ...[/quote]
Indeed "the same drawing". I had already got those from the score table.
All the best, --stolfi
18-11-2025, 09:17 AM
(17-11-2025, 05:37 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Thanks! But, unless I again got the wrong page numbers, I don't see those pairs with You are not allowed to view links. Register or Login to view. as definitely "the same drawing".
Indeed, so they are listed as 'some similarity'.
By the way, one of my unprovable hypotheses is that we only have part of all the planned full-page herbal drawings. If there is ever a translation, we might be able to say more about that.
18-11-2025, 12:03 PM
(18-11-2025, 09:17 AM)ReneZ Wrote: You are not allowed to view links. Register or Login to view.By the way, one of my unprovable hypotheses is that we only have part of all the planned full-page herbal drawings. If there is ever a translation, we might be able to say more about that.
For my part, I suspect that we have only a small fraction of the original Pharma pages. Perhaps in the latter there would be good matches for almost all Herbal pages..,
All the best, --stolfi
18-11-2025, 10:20 PM
These are the pairs I have got so far:
The fields in each entry are: first page's f-name, indices of plant in the page, crop coordinates, ditto for the second page, kind of match, and observations.
The indices are (row,plant) counting from 1, from top and left, for Pharma pages; or "None" for the single plant in Herbal pages.
Page f89v3 is traditionally but inconsistently named f89v2.
The crop coordinates are as in the ImageMagick "convert" tool and refer to the BL 2014 images as downloaded from the Beinecke site.
The kind of match is 1 for good Herbal-Pharma, 2 for good Herbal-Herbal, and 3 for proposed but unconvincing matches. Only two of the latter were included, but there are many more like them.
Hope it helps, --stolfi
Code:
plant_pairs = (
("f1v", None, "2200x3300+0480+0348", "f102r1", (3,2), "1300x1170+3784+2172", 1, "Root, leaves; NO flower."),
("f18v", None, "2200x3400+0500+0310", "f102r2", (3,1), "1600x1100+5240+2470", 1, "Root; NO leaves, flower."),
("f19r", None, "2100x3300+0316+0400", "f102v1", (2,3), "0800x0800+1150+1780", 1, "Root, leaves (type 1 only); NO flowers."),
("f23r", None, "2260x3100+0200+0320", "f102r2", (3,2), "1100x1000+6800+2480", 1, "Root; NO leaves, flowers."),
("f37v", None, "2100x3500+0685+0200", "f102r1", (3,1), "1200x1500+2850+2080", 1, "Root, leaves; NO flower"),
("f32v", None, "2100x3600+0600+0140", "f102r2", (1,2), "1600x1100+6460+0320", 1, "Root, leaves, flower."),
("f47v", None, "2300x3500+0580+0210", "f102r2", (1,1), "1400x0800+5460+0240", 1, "Root, leaves; NO flower."),
("f48r", None, "2200x3200+0249+0162", "f89v3", (3,4), "1300x1300+2700+2440", 1, "Root, leaves; NO flowers."),
("f48v", None, "2100x3300+0585+0330", "f89v3", (1,5), "1000x1300+3000+0100", 1, "Root, leaves; NO flowers."),
("f39r", None, "2300x3400+0108+0292", "f95r2", None, "2200x3560+5340+0128", 2, "Root, leaves; NOT flower."),
("f51r", None, "1700x3300+0850+0400", "f99r", (3,1), "0200x0400+0490+1050", 3, "Root; NO leaves, flowers."),
("f90v1", None, "2300x3660+0500+0060", "f99v", (3,2), "0500x0500+0990+1600", 3, "Root; NO leaves, flowers."),
)The fields in each entry are: first page's f-name, indices of plant in the page, crop coordinates, ditto for the second page, kind of match, and observations.
The indices are (row,plant) counting from 1, from top and left, for Pharma pages; or "None" for the single plant in Herbal pages.
Page f89v3 is traditionally but inconsistently named f89v2.
The crop coordinates are as in the ImageMagick "convert" tool and refer to the BL 2014 images as downloaded from the Beinecke site.
The kind of match is 1 for good Herbal-Pharma, 2 for good Herbal-Herbal, and 3 for proposed but unconvincing matches. Only two of the latter were included, but there are many more like them.
Hope it helps, --stolfi