Skip to main content
. 2022 Jun 4;12(11):1453. doi: 10.3390/ani12111453

Table 1.

Reference summary from previous cattle muzzle identification studies.

Cattle Type Image Size (Pixels) Image Type Restrained Cattle Counts Images per Cattle Total Images Identification Method Accuracy (%) Processing Time (ms/Image) Reference
Dairy Printed Y 6 Manual [16]
Printed Y 200 Manual [17]
Printed Y 65 Manual [18]
Beef 256 × 256 Grayscale Y 43 DIP 46.5 [31]
Beef 320 × 240 Printed Y 29 10 290 ML 98.9 [12]
Beef 200 × 200 Grayscale 8 10 80 DIP 90.0 [32]
Grayscale 15 7 105 DIP 93.3 37–879 [15]
Beef Printed Y 20 8 160 DIP 98.3 [20]
Grayscale 53 20 1060 DIP [19]
Beef 300 × 400 Grayscale 31 7 217 ML 99.5 [33]
RGB 28 20 560 ML 100.0 [25]
RGB 52 20 1040 ML 96.0 [24]
Beef RGB N 14 5 70 DIP 100.0 [21]
Beef 300 × 400 Grayscale 31 7 217 ML 99.5 [34]
Beef 300 × 400 Grayscale 31 7 217 ML 99.5 48–1362 [23]
Beef RGB 52 6 312 ML 96.4 [35]
Dairy 400 × 400 RGB 500 10 5000 DIP 93.9 [36]
Dairy 200 × 200 RGB 500 10 5000 ML 94.9 [37]
Dairy 200 × 200 RGB 500 10 5000 DL 98.9 [27]
Dairy 200 × 200 RGB 500 10 5000 ML 93.9 [38]
Dairy RGB N 15 7 105 ML 93.0 368–1193 [30]
Beef RGB Y 60 5–10 460 DIP 98.1 [39]
Beef RGB 45 20 900 ML 96.5 [40]
Beef RGB Y 431 1600 ML 95.0 [41]
Dairy 200 × 200 RGB 400 10 4000 DL 98.9 [28]
Beef 1024 × 1024 RGB Y 300 2900 DL 99.1 [29]
Dairy 64 × 64 RGB 186 5 930 ML 83.4 [13]

Note: ‘’ indicates that information was not provided. DIP, digital image processing; ML, machine learning; DL, deep learning. Cattle species include beef cattle and dairy cattle. Image type is categorized as printed (samples are obtained from a direct compress with cattle noses and then scanned or photographed to form electronic images), grayscale with one-channel data captured directly from cameras, and RGB with three-channel (red, green, and blue) data. ‘Y’ indicates that the animal was restrained during data collection, while ‘N’ indicates that it was not.