Oval Detection Report

2020 General Election, Volusia County, FL

AuditEngine version:'v2.32.X'

Run Date: 2026-11-11 15:58:42.542123 Copyright (c) Citizens' Oversight

Return to Master Report Index

Settings in effect for this report

setting value note
vendor 'ES&S'
target_side 'left' should drive the default option/ocr clip region direction in vendor.py's layout_params for a right-side-target ballot, same as option_clip_region already does -- ocr_clip_region has no such default yet, see vendor.py
ml_min_detection_probability_pct 80 positions scoring below this are never detected at all
grid_rows_to_drop 2 last N y-grid rows excluded from candidate positions entirely -- no in-code default, vendor-specific unless a job overrides it
ml_target_exclude_regions [{'x': 0, 'y': 0, 'w': 1700, 'h': 460, 'page': 1}]
ml_target_eval_w x ml_target_eval_h 72 x 32 classifier's own evaluation-clip size
target_offset_x/y (page 1) 0, 0
target_offset_x_p1/y_p1 (page 2+) 0, 0
mapping_ocr_detect_only False
mapping_ocr_match_only True map_option_names_ocr()'s own two-flag split -- affects whether map/ml_detected_targets.csv reflects a real detection pass or a reused one, not this report's own generation

Summary

175 of 177 styles have detected oval count exactly matching the EIF-expected count.

2 styles do NOT match ('diff': positive = more detected than expected, negative = fewer) -- also saved as map/oval_count_mismatches.csv:

style_num expected detected diff
085 62 55 -7
099 62 46 -16

X-bin Column Analysis

Total x-bins discovered: 21 (real columns: 6, outliers/noise: 15)

Real Columns (common x-bins, ≥50% style coverage)

x_avg coverage num_styles num_dets style_count
107.3 100.0% 177 3743 177
107.8 100.0% 165 378 165
622.9 89.8% 159 3140 159
1140.1 89.8% 159 636 159
625.4 96.4% 159 636 159
1142.2 96.4% 159 953 159

Outlier/Noise X-bins (<50% coverage)

x_avg coverage num_styles num_dets affected_styles
819.9 5.5% 9 9
236.6 4.0% 7 7
690.1 3.6% 6 7
366.2 3.0% 5 5
1205.3 1.8% 3 3
1593.3 1.8% 3 3
171.0 1.1% 2 3
1080.0 1.2% 2 2
428.0 0.6% 1 1
685.0 0.6% 1 1
879.0 0.6% 1 1
1400.0 0.6% 1 1
558.0 0.6% 1 2
951.0 0.6% 1 1
1012.0 0.6% 1 1

Full Oval List for Mismatched Styles

Every detected oval for a style whose count still doesn't match expected, after xbin filtering -- oval match % is the classifier's own probability; bin membership % is the share of the job's styles sharing that oval's x-bin (its real-column confidence). Neither alone identifies which oval is the extra/missing one here -- that's exactly why the style is still unresolved -- but seeing both together may reveal a pattern (e.g. a consistently lower match % or a borderline bin %) the filter isn't using yet.

Style 085 (expected 62, detected 55, diff -7)

page x y oval_match_pct bin_x_avg bin_membership_pct common_bin
1 106 475 99.8% 107.3 100.0% True
1 106 557 99.5% 107.3 100.0% True
1 106 640 99.8% 107.3 100.0% True
1 106 722 99.7% 107.3 100.0% True
1 106 804 99.7% 107.3 100.0% True
1 106 885 100.0% 107.3 100.0% True
1 106 967 98.6% 107.3 100.0% True
1 106 1048 99.6% 107.3 100.0% True
1 106 1294 99.3% 107.3 100.0% True
1 106 1335 98.1% 107.3 100.0% True
1 106 1376 98.7% 107.3 100.0% True
1 106 1580 96.4% 107.3 100.0% True
1 106 1621 98.8% 107.3 100.0% True
1 106 1825 92.0% 107.3 100.0% True
1 106 1866 93.6% 107.3 100.0% True
1 106 1906 96.1% 107.3 100.0% True
1 106 2152 92.8% 107.3 100.0% True
1 106 2193 98.3% 107.3 100.0% True
1 106 2479 97.8% 107.3 100.0% True
1 106 2764 85.0% 107.3 100.0% True
1 106 2805 98.4% 107.3 100.0% True
1 106 3009 99.3% 107.3 100.0% True
1 106 3049 99.7% 107.3 100.0% True
1 623 598 89.3% 622.9 89.8% True
1 623 640 93.2% 622.9 89.8% True
1 623 844 99.5% 622.9 89.8% True
1 623 885 98.8% 622.9 89.8% True
1 623 1131 84.8% 622.9 89.8% True
1 623 1376 80.3% 622.9 89.8% True
1 623 1539 94.3% 622.9 89.8% True
1 623 1580 94.4% 622.9 89.8% True
1 623 1784 96.2% 622.9 89.8% True
1 623 2029 92.0% 622.9 89.8% True
1 623 2274 89.8% 622.9 89.8% True
1 623 2520 88.0% 622.9 89.8% True
1 623 2764 93.4% 622.9 89.8% True
1 623 2805 91.5% 622.9 89.8% True
1 623 2969 80.2% 622.9 89.8% True
1 623 3009 99.0% 622.9 89.8% True
1 1141 1294 93.0% 1140.1 89.8% True
1 1141 1335 92.7% 1140.1 89.8% True
1 1141 3131 99.6% 1140.1 89.8% True
1 1141 3173 98.8% 1140.1 89.8% True
2 107 2272 91.0% 107.3 100.0% True
2 107 2312 96.4% 107.3 100.0% True
2 624 2107 97.5% 622.9 89.8% True
2 624 2148 80.2% 622.9 89.8% True
2 624 2884 100.0% 622.9 89.8% True
2 624 2925 89.5% 622.9 89.8% True
2 1141 963 99.9% 1140.1 89.8% True
2 1141 1003 94.8% 1140.1 89.8% True
2 1141 1943 99.9% 1140.1 89.8% True
2 1141 2025 99.8% 1140.1 89.8% True
2 1141 3089 99.8% 1140.1 89.8% True
2 1141 3170 99.9% 1140.1 89.8% True

Style 099 (expected 62, detected 46, diff -16)

page x y oval_match_pct bin_x_avg bin_membership_pct common_bin
1 107 475 100.0% 107.3 100.0% True
1 107 558 99.5% 107.3 100.0% True
1 107 640 98.7% 107.3 100.0% True
1 107 720 99.9% 107.3 100.0% True
1 107 802 97.8% 107.3 100.0% True
1 107 884 95.7% 107.3 100.0% True
1 107 966 91.2% 107.3 100.0% True
1 107 1047 95.4% 107.3 100.0% True
1 107 1293 80.1% 107.3 100.0% True
1 107 1375 95.5% 107.3 100.0% True
1 107 1824 94.7% 107.3 100.0% True
1 107 1866 91.9% 107.3 100.0% True
1 107 1906 90.3% 107.3 100.0% True
1 107 2151 97.1% 107.3 100.0% True
1 107 2438 93.9% 107.3 100.0% True
1 107 2479 81.2% 107.3 100.0% True
1 107 2764 98.1% 107.3 100.0% True
1 107 3009 99.8% 107.3 100.0% True
1 107 3051 99.4% 107.3 100.0% True
1 624 598 99.2% 622.9 89.8% True
1 624 640 95.7% 622.9 89.8% True
1 624 843 99.3% 622.9 89.8% True
1 624 884 90.0% 622.9 89.8% True
1 624 1089 95.3% 622.9 89.8% True
1 624 1129 80.4% 622.9 89.8% True
1 624 1335 83.6% 622.9 89.8% True
1 624 1375 85.7% 622.9 89.8% True
1 624 1538 94.6% 622.9 89.8% True
1 624 1742 98.8% 622.9 89.8% True
1 624 1988 96.7% 622.9 89.8% True
1 624 2233 98.1% 622.9 89.8% True
1 624 2764 86.9% 622.9 89.8% True
1 624 3009 95.5% 622.9 89.8% True
1 1141 3133 98.3% 1140.1 89.8% True
1 1141 3173 99.6% 1140.1 89.8% True
2 106 2273 98.7% 107.3 100.0% True
2 106 2314 99.5% 107.3 100.0% True
2 623 2110 97.7% 622.9 89.8% True
2 623 2150 94.7% 622.9 89.8% True
2 623 2886 84.8% 622.9 89.8% True
2 623 2927 96.6% 622.9 89.8% True
2 1141 1007 90.5% 1140.1 89.8% True
2 1141 1946 83.2% 1140.1 89.8% True
2 1141 2028 96.2% 1140.1 89.8% True
2 1141 3090 98.0% 1140.1 89.8% True
2 1141 3173 97.8% 1140.1 89.8% True

Full snapgrid (every 'top'/'lft' line, edge to edge) in thin red, and the oval classifier's own evaluation-clip boundary at EVERY candidate grid position (not just accepted detections, keepout regions already excluded) in thin green -- lets you check grid registration against the page's real timing marks and spot an eval window overlapping neighboring ballot content directly. On top of that: for every style with detections in map/ml_detected_targets.csv, a bold red box + 'x,y,confidence%' around every detected oval, a bold blue box around the region that will be sent to the AI/OCR matching stage, and a bold green box previewing the eventual crop boundary. 354 images across 177 styles.

001

001-mldetect-redlined-1.png


001-mldetect-redlined-2.png


002

002-mldetect-redlined-1.png


002-mldetect-redlined-2.png


003

003-mldetect-redlined-1.png


003-mldetect-redlined-2.png


004

004-mldetect-redlined-1.png


004-mldetect-redlined-2.png


005

005-mldetect-redlined-1.png


005-mldetect-redlined-2.png


006

006-mldetect-redlined-1.png


006-mldetect-redlined-2.png


007

007-mldetect-redlined-1.png


007-mldetect-redlined-2.png


008

008-mldetect-redlined-1.png


008-mldetect-redlined-2.png


009

009-mldetect-redlined-1.png


009-mldetect-redlined-2.png


010

010-mldetect-redlined-1.png


010-mldetect-redlined-2.png


011

011-mldetect-redlined-1.png


011-mldetect-redlined-2.png


013

013-mldetect-redlined-1.png


013-mldetect-redlined-2.png


014

014-mldetect-redlined-1.png


014-mldetect-redlined-2.png


015

015-mldetect-redlined-1.png


015-mldetect-redlined-2.png


016

016-mldetect-redlined-1.png


016-mldetect-redlined-2.png


017

017-mldetect-redlined-1.png


017-mldetect-redlined-2.png


018

018-mldetect-redlined-1.png


018-mldetect-redlined-2.png


019

019-mldetect-redlined-1.png


019-mldetect-redlined-2.png


020

020-mldetect-redlined-1.png


020-mldetect-redlined-2.png


021

021-mldetect-redlined-1.png


021-mldetect-redlined-2.png


022

022-mldetect-redlined-1.png


022-mldetect-redlined-2.png


023

023-mldetect-redlined-1.png


023-mldetect-redlined-2.png


024

024-mldetect-redlined-1.png


024-mldetect-redlined-2.png


025

025-mldetect-redlined-1.png


025-mldetect-redlined-2.png


026

026-mldetect-redlined-1.png


026-mldetect-redlined-2.png


027

027-mldetect-redlined-1.png


027-mldetect-redlined-2.png


028

028-mldetect-redlined-1.png


028-mldetect-redlined-2.png


029

029-mldetect-redlined-1.png


029-mldetect-redlined-2.png


030

030-mldetect-redlined-1.png


030-mldetect-redlined-2.png


031

031-mldetect-redlined-1.png


031-mldetect-redlined-2.png


032

032-mldetect-redlined-1.png


032-mldetect-redlined-2.png


033

033-mldetect-redlined-1.png


033-mldetect-redlined-2.png


034

034-mldetect-redlined-1.png


034-mldetect-redlined-2.png


035

035-mldetect-redlined-1.png


035-mldetect-redlined-2.png


036

036-mldetect-redlined-1.png


036-mldetect-redlined-2.png


037

037-mldetect-redlined-1.png


037-mldetect-redlined-2.png


038

038-mldetect-redlined-1.png


038-mldetect-redlined-2.png


039

039-mldetect-redlined-1.png


039-mldetect-redlined-2.png


040

040-mldetect-redlined-1.png


040-mldetect-redlined-2.png


041

041-mldetect-redlined-1.png


041-mldetect-redlined-2.png


042

042-mldetect-redlined-1.png


042-mldetect-redlined-2.png


043

043-mldetect-redlined-1.png


043-mldetect-redlined-2.png


044

044-mldetect-redlined-1.png


044-mldetect-redlined-2.png


045

045-mldetect-redlined-1.png


045-mldetect-redlined-2.png


046

046-mldetect-redlined-1.png


046-mldetect-redlined-2.png


047

047-mldetect-redlined-1.png


047-mldetect-redlined-2.png


048

048-mldetect-redlined-1.png


048-mldetect-redlined-2.png


049

049-mldetect-redlined-1.png


049-mldetect-redlined-2.png


050

050-mldetect-redlined-1.png


050-mldetect-redlined-2.png


051

051-mldetect-redlined-1.png


051-mldetect-redlined-2.png


052

052-mldetect-redlined-1.png


052-mldetect-redlined-2.png


053

053-mldetect-redlined-1.png


053-mldetect-redlined-2.png


054

054-mldetect-redlined-1.png


054-mldetect-redlined-2.png


055

055-mldetect-redlined-1.png


055-mldetect-redlined-2.png


056

056-mldetect-redlined-1.png


056-mldetect-redlined-2.png


057

057-mldetect-redlined-1.png


057-mldetect-redlined-2.png


058

058-mldetect-redlined-1.png


058-mldetect-redlined-2.png


059

059-mldetect-redlined-1.png


059-mldetect-redlined-2.png


060

060-mldetect-redlined-1.png


060-mldetect-redlined-2.png


061

061-mldetect-redlined-1.png


061-mldetect-redlined-2.png


062

062-mldetect-redlined-1.png


062-mldetect-redlined-2.png


063

063-mldetect-redlined-1.png


063-mldetect-redlined-2.png


064

064-mldetect-redlined-1.png


064-mldetect-redlined-2.png


065

065-mldetect-redlined-1.png


065-mldetect-redlined-2.png


066

066-mldetect-redlined-1.png


066-mldetect-redlined-2.png


067

067-mldetect-redlined-1.png


067-mldetect-redlined-2.png


068

068-mldetect-redlined-1.png


068-mldetect-redlined-2.png


069

069-mldetect-redlined-1.png


069-mldetect-redlined-2.png


070

070-mldetect-redlined-1.png


070-mldetect-redlined-2.png


071

071-mldetect-redlined-1.png


071-mldetect-redlined-2.png


072

072-mldetect-redlined-1.png


072-mldetect-redlined-2.png


073

073-mldetect-redlined-1.png


073-mldetect-redlined-2.png


074

074-mldetect-redlined-1.png


074-mldetect-redlined-2.png


075

075-mldetect-redlined-1.png


075-mldetect-redlined-2.png


076

076-mldetect-redlined-1.png


076-mldetect-redlined-2.png


077

077-mldetect-redlined-1.png


077-mldetect-redlined-2.png


078

078-mldetect-redlined-1.png


078-mldetect-redlined-2.png


079

079-mldetect-redlined-1.png


079-mldetect-redlined-2.png


080

080-mldetect-redlined-1.png


080-mldetect-redlined-2.png


081

081-mldetect-redlined-1.png


081-mldetect-redlined-2.png


082

082-mldetect-redlined-1.png


082-mldetect-redlined-2.png


083

083-mldetect-redlined-1.png


083-mldetect-redlined-2.png


084

084-mldetect-redlined-1.png


084-mldetect-redlined-2.png


085

085-mldetect-redlined-1.png


085-mldetect-redlined-2.png


086

086-mldetect-redlined-1.png


086-mldetect-redlined-2.png


087

087-mldetect-redlined-1.png


087-mldetect-redlined-2.png


088

088-mldetect-redlined-1.png


088-mldetect-redlined-2.png


089

089-mldetect-redlined-1.png


089-mldetect-redlined-2.png


090

090-mldetect-redlined-1.png


090-mldetect-redlined-2.png


091

091-mldetect-redlined-1.png


091-mldetect-redlined-2.png


092

092-mldetect-redlined-1.png


092-mldetect-redlined-2.png


093

093-mldetect-redlined-1.png


093-mldetect-redlined-2.png


094

094-mldetect-redlined-1.png


094-mldetect-redlined-2.png


096

096-mldetect-redlined-1.png


096-mldetect-redlined-2.png


099

099-mldetect-redlined-1.png


099-mldetect-redlined-2.png


101

101-mldetect-redlined-1.png


101-mldetect-redlined-2.png


102

102-mldetect-redlined-1.png


102-mldetect-redlined-2.png


104

104-mldetect-redlined-1.png


104-mldetect-redlined-2.png


106

106-mldetect-redlined-1.png


106-mldetect-redlined-2.png


107

107-mldetect-redlined-1.png


107-mldetect-redlined-2.png


108

108-mldetect-redlined-1.png


108-mldetect-redlined-2.png


109

109-mldetect-redlined-1.png


109-mldetect-redlined-2.png


110

110-mldetect-redlined-1.png


110-mldetect-redlined-2.png


111

111-mldetect-redlined-1.png


111-mldetect-redlined-2.png


112

112-mldetect-redlined-1.png


112-mldetect-redlined-2.png


113

113-mldetect-redlined-1.png


113-mldetect-redlined-2.png


114

114-mldetect-redlined-1.png


114-mldetect-redlined-2.png


115

115-mldetect-redlined-1.png


115-mldetect-redlined-2.png


116

116-mldetect-redlined-1.png


116-mldetect-redlined-2.png


117

117-mldetect-redlined-1.png


117-mldetect-redlined-2.png


118

118-mldetect-redlined-1.png


118-mldetect-redlined-2.png


119

119-mldetect-redlined-1.png


119-mldetect-redlined-2.png


120

120-mldetect-redlined-1.png


120-mldetect-redlined-2.png


122

122-mldetect-redlined-1.png


122-mldetect-redlined-2.png


123

123-mldetect-redlined-1.png


123-mldetect-redlined-2.png


124

124-mldetect-redlined-1.png


124-mldetect-redlined-2.png


125

125-mldetect-redlined-1.png


125-mldetect-redlined-2.png


126

126-mldetect-redlined-1.png


126-mldetect-redlined-2.png


128

128-mldetect-redlined-1.png


128-mldetect-redlined-2.png


129

129-mldetect-redlined-1.png


129-mldetect-redlined-2.png


130

130-mldetect-redlined-1.png


130-mldetect-redlined-2.png


132

132-mldetect-redlined-1.png


132-mldetect-redlined-2.png


133

133-mldetect-redlined-1.png


133-mldetect-redlined-2.png


134

134-mldetect-redlined-1.png


134-mldetect-redlined-2.png


135

135-mldetect-redlined-1.png


135-mldetect-redlined-2.png


136

136-mldetect-redlined-1.png


136-mldetect-redlined-2.png


137

137-mldetect-redlined-1.png


137-mldetect-redlined-2.png


138

138-mldetect-redlined-1.png


138-mldetect-redlined-2.png


139

139-mldetect-redlined-1.png


139-mldetect-redlined-2.png


140

140-mldetect-redlined-1.png


140-mldetect-redlined-2.png


141

141-mldetect-redlined-1.png


141-mldetect-redlined-2.png


142

142-mldetect-redlined-1.png


142-mldetect-redlined-2.png


143

143-mldetect-redlined-1.png


143-mldetect-redlined-2.png


144

144-mldetect-redlined-1.png


144-mldetect-redlined-2.png


145

145-mldetect-redlined-1.png


145-mldetect-redlined-2.png


146

146-mldetect-redlined-1.png


146-mldetect-redlined-2.png


147

147-mldetect-redlined-1.png


147-mldetect-redlined-2.png


148

148-mldetect-redlined-1.png


148-mldetect-redlined-2.png


149

149-mldetect-redlined-1.png


149-mldetect-redlined-2.png


150

150-mldetect-redlined-1.png


150-mldetect-redlined-2.png


152

152-mldetect-redlined-1.png


152-mldetect-redlined-2.png


153

153-mldetect-redlined-1.png


153-mldetect-redlined-2.png


154

154-mldetect-redlined-1.png


154-mldetect-redlined-2.png


155

155-mldetect-redlined-1.png


155-mldetect-redlined-2.png


156

156-mldetect-redlined-1.png


156-mldetect-redlined-2.png


157

157-mldetect-redlined-1.png


157-mldetect-redlined-2.png


158

158-mldetect-redlined-1.png


158-mldetect-redlined-2.png


159

159-mldetect-redlined-1.png


159-mldetect-redlined-2.png


160

160-mldetect-redlined-1.png


160-mldetect-redlined-2.png


161

161-mldetect-redlined-1.png


161-mldetect-redlined-2.png


162

162-mldetect-redlined-1.png


162-mldetect-redlined-2.png


163

163-mldetect-redlined-1.png


163-mldetect-redlined-2.png


164

164-mldetect-redlined-1.png


164-mldetect-redlined-2.png


165

165-mldetect-redlined-1.png


165-mldetect-redlined-2.png


166

166-mldetect-redlined-1.png


166-mldetect-redlined-2.png


167

167-mldetect-redlined-1.png


167-mldetect-redlined-2.png


168

168-mldetect-redlined-1.png


168-mldetect-redlined-2.png


169

169-mldetect-redlined-1.png


169-mldetect-redlined-2.png


203

203-mldetect-redlined-1.png


203-mldetect-redlined-2.png


204

204-mldetect-redlined-1.png


204-mldetect-redlined-2.png


205

205-mldetect-redlined-1.png


205-mldetect-redlined-2.png


206

206-mldetect-redlined-1.png


206-mldetect-redlined-2.png


211

211-mldetect-redlined-1.png


211-mldetect-redlined-2.png


316

316-mldetect-redlined-1.png


316-mldetect-redlined-2.png


317

317-mldetect-redlined-1.png


317-mldetect-redlined-2.png


318

318-mldetect-redlined-1.png


318-mldetect-redlined-2.png


321

321-mldetect-redlined-1.png


321-mldetect-redlined-2.png


322

322-mldetect-redlined-1.png


322-mldetect-redlined-2.png


323

323-mldetect-redlined-1.png


323-mldetect-redlined-2.png


324

324-mldetect-redlined-1.png


324-mldetect-redlined-2.png


325

325-mldetect-redlined-1.png


325-mldetect-redlined-2.png


331

331-mldetect-redlined-1.png


331-mldetect-redlined-2.png


332

332-mldetect-redlined-1.png


332-mldetect-redlined-2.png


333

333-mldetect-redlined-1.png


333-mldetect-redlined-2.png


334

334-mldetect-redlined-1.png


334-mldetect-redlined-2.png


335

335-mldetect-redlined-1.png


335-mldetect-redlined-2.png


338

338-mldetect-redlined-1.png


338-mldetect-redlined-2.png


End of Oval Detection Report

AuditEngine version:'v2.32.X'

Run Date: 2026-11-11 15:58:42.542645 Copyright (c) Citizens' Oversight

Return to Master Report Index