关于「显著性:一瞥之内,谁先被挑中」,《服饰图案远距辨识与再认》给出的定量结论是什么?
结论:D1 是 26 厘米、剪影对比 9.13、闭合度 0.9 的徽记,对应第三章白墨半版印黑衣。它在一米外的示意概率是 0.877,五米 0.590,十米 0.418,十五米 0.324。D2 尺寸与闭合度相同,对比换成玫红印红的 1.36,十米外只剩 0.121。D3 把尺寸缩到 12 厘米,十米外 0.249。D4 把闭合度降到 0.3,十米外 0.226。D5 是 40 厘米的大图案,对比 4.25、闭合度 0.6,十米外 0.266,大并没有换来更多。
数据口径:书中参数为示意,全部计算由书中程序生成。 本书不是商标或著作权法律意见,示意图案全部为作者自拟几何形,不点名评价任何真实品牌,不写任何潜意识影响宣称,街头测试只记录匿名对错,不采集人脸与个人资料。
对照数据表
下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.121, 0.226, 0.249, 0.266, 0.3, 0.324
| A. illustrative capture chance by distance | |
| design | 1m 3m 5m 10m 15m |
| D1 26cm CR9.13 k0.9 | 0.877 0.705 0.590 0.418 0.324 |
| D2 26cm CR1.36 k0.9 | 0.578 0.314 0.216 0.121 0.084 |
| D3 12cm CR9.13 k0.9 | 0.768 0.525 0.399 0.249 0.181 |
| D4 26cm CR9.13 k0.3 | 0.744 0.493 0.369 0.226 0.163 |
| D5 40cm CR4.25 k0.6 | 0.782 0.547 0.420 0.266 0.195 |
| B. log-odds gained at 10 m by each design move | |
| double the size: +0.693 | |
| double the contrast: +0.602 | |
| closure 0.3 to 0.9: +0.900 | |
| C. D1 at 10 m: P | 0.418; at 15 m the size needed to match: 39.0 cm |
查看完整验算程序
# Chapter 4. p43: an illustrative logistic for a glance-capture chance.
# Model (illustrative parameters, not fitted to any data):
# logit P = B0 + B1 ln(theta_deg) + B2 log10(CR) + B3 k
# theta = angular size of the emblem silhouette, CR = its silhouette
# contrast ratio (p33 analogy), k = closure score in [0, 1] (chapter 5).
# P is read as "the emblem is fixated within the first glance", a
# quantity to be estimated by the reader's own tests, never assumed.
import math
B0, B1, B2, B3 = -4.0, 1.0, 2.0, 1.5
DIST = (1.0, 3.0, 5.0, 10.0, 15.0)
def theta_deg(size_cm, d_m):
return math.degrees(2.0 * math.atan((size_cm / 100.0) / (2.0 * d_m)))
def p_capture(size_cm, d_m, cr, k):
z = B0 + B1 * math.log(theta_deg(size_cm, d_m)) \
+ B2 * math.log10(cr) + B3 * k
return 1.0 / (1.0 + math.exp(-z))
DESIGNS = (
("D1 26cm CR9.13 k0.9", 26.0, 9.13, 0.9),
("D2 26cm CR1.36 k0.9", 26.0, 1.36, 0.9),
("D3 12cm CR9.13 k0.9", 12.0, 9.13, 0.9),
("D4 26cm CR9.13 k0.3", 26.0, 9.13, 0.3),
("D5 40cm CR4.25 k0.6", 40.0, 4.25, 0.6),
)
if __name__ == "__main__":
print("A. illustrative capture chance by distance")
print(" design " + "".join("{:>6.0f}m".format(d)
for d in DIST))
for name, size, cr, k in DESIGNS:
print(" {:<21} ".format(name) + "".join(
"{:>7.3f}".format(p_capture(size, d, cr, k)) for d in DIST))
print("B. log-odds gained at 10 m by each design move")
print(" double the size: {:+.3f}".format(B1 * math.log(2.0)))
print(" double the contrast: {:+.3f}".format(B2 * math.log10(2.0)))
print(" closure 0.3 to 0.9: {:+.3f}".format(B3 * 0.6))
p = p_capture(26.0, 10.0, 9.13, 0.9)
print("C. D1 at 10 m: P = {:.3f}; at 15 m the size needed to match: "
"{:.1f} cm".format(p, 26.0 * 15.0 / 10.0))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《服饰图腾完形视知觉捕获、远距高对比辨识与街头视觉模因驻留母典》第 4 章「显著性:一瞥之内,谁先被挑中」(版本 1.0.0). 5D Library. DOI 待发布
DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-390 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026390,
author = {Fan, Yafu},
title = {Apparel Graphic Gestalt Capture, Long-Range Contrast Legibility and Streetwear Visual Memetic Retention},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}