■ ONNX와 YOLOv5를 사용해 이미지 객체를 인식하는 방법을 보여준다.
[소스 이미지]
[결과 이미지]
▶ RectangleExtension.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 |
using System.Drawing; namespace TestProject { /// <summary> /// 사각형 확장 /// </summary> public static class RectangleExtension { //////////////////////////////////////////////////////////////////////////////////////////////////// Method ////////////////////////////////////////////////////////////////////////////////////////// Static //////////////////////////////////////////////////////////////////////////////// Public #region 영역 구하기 - GetArea(rectangle) /// <summary> /// 영역 구하기 /// </summary> /// <param name="rectangle">사각형</param> /// <returns>영역</returns> public static float GetArea(this RectangleF rectangle) { return rectangle.Width * rectangle.Height; } #endregion } } |
▶ YOLOModel.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 |
using System.Collections.Generic; namespace TestProject { /// <summary> /// YOLO 모델 /// </summary> public abstract class YOLOModel { //////////////////////////////////////////////////////////////////////////////////////////////////// Property ////////////////////////////////////////////////////////////////////////////////////////// Public #region 너비 - Width /// <summary> /// 너비 /// </summary> public abstract int Width { get; set; } #endregion #region 높이 - Height /// <summary> /// 높이 /// </summary> public abstract int Height { get; set; } #endregion #region 깊이 - Depth /// <summary> /// 깊이 /// </summary> public abstract int Depth { get; set; } #endregion #region 차원 - Dimension /// <summary> /// 차원 /// </summary> public abstract int Dimension { get; set; } #endregion #region 스트라이드 배열 - StrideArray /// <summary> /// 스트라이드 배열 /// </summary> public abstract float[] StrideArray { get; set; } #endregion #region 앵커 배열 - AnchorArray /// <summary> /// 앵커 배열 /// </summary> public abstract float[][][] AnchorArray { get; set; } #endregion #region 도형 배열 - ShapeArray /// <summary> /// 도형 배열 /// </summary> public abstract int[] ShapeArray { get; set; } #endregion #region 신뢰도 - Confidence /// <summary> /// 신뢰도 /// </summary> public abstract float Confidence { get; set; } #endregion #region 다중 신뢰도 - MultipleConfidence /// <summary> /// 다중 신뢰도 /// </summary> public abstract float MultipleConfidence { get; set; } #endregion #region 오버랩 - Overlap /// <summary> /// 오버랩 /// </summary> public abstract float Overlap { get; set; } #endregion #region 출력 배열 - OutputArray /// <summary> /// 출력 배열 /// </summary> public abstract string[] OutputArray { get; set; } #endregion #region 레이블 리스트 - LabelList /// <summary> /// 레이블 리스트 /// </summary> public abstract List<YOLOLabel> LabelList { get; set; } #endregion #region 탐지 사용 여부 - UseDetect /// <summary> /// 탐지 사용 여부 /// </summary> public abstract bool UseDetect { get; set; } #endregion } } |
▶ YOLOCOCOP5Model.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 |
using System.Collections.Generic; namespace TestProject { /// <summary> /// YOLO COCO P5 모델 /// </summary> public class YOLOCOCOP5Model : YOLOModel { //////////////////////////////////////////////////////////////////////////////////////////////////// Property ////////////////////////////////////////////////////////////////////////////////////////// Public #region 너비 - Width /// <summary> /// 너비 /// </summary> public override int Width { get; set; } = 640; #endregion #region 높이 - Height /// <summary> /// 높이 /// </summary> public override int Height { get; set; } = 640; #endregion #region 깊이 - Depth /// <summary> /// 깊이 /// </summary> public override int Depth { get; set; } = 3; #endregion #region 차원 - Dimension /// <summary> /// 차원 /// </summary> public override int Dimension { get; set; } = 85; #endregion #region 스트라이드 배열 - StrideArray /// <summary> /// 스트라이드 배열 /// </summary> public override float[] StrideArray { get; set; } = new float[] { 8, 16, 32 }; #endregion #region 앵커 배열 - AnchorArray /// <summary> /// 앵커 배열 /// </summary> public override float[][][] AnchorArray { get; set; } = new float[][][] { new float[][] { new float[] { 010, 13 }, new float[] { 016, 030 }, new float[] { 033, 023 } }, new float[][] { new float[] { 030, 61 }, new float[] { 062, 045 }, new float[] { 059, 119 } }, new float[][] { new float[] { 116, 90 }, new float[] { 156, 198 }, new float[] { 373, 326 } } }; #endregion #region 도형 배열 - ShapeArray /// <summary> /// 도형 배열 /// </summary> public override int[] ShapeArray { get; set; } = new int[] { 80, 40, 20 }; #endregion #region 신뢰도 - Confidence /// <summary> /// 신뢰도 /// </summary> public override float Confidence { get; set; } = 0.20f; #endregion #region 다중 신뢰도 - MultipleConfidence /// <summary> /// 다중 신뢰도 /// </summary> public override float MultipleConfidence { get; set; } = 0.25f; #endregion #region 오버랩 - Overlap /// <summary> /// 오버랩 /// </summary> public override float Overlap { get; set; } = 0.45f; #endregion #region 출력 배열 - OutputArray /// <summary> /// 출력 배열 /// </summary> public override string[] OutputArray { get; set; } = new[] { "output" }; #endregion #region 레이블 리스트 - LabelList /// <summary> /// 레이블 리스트 /// </summary> public override List<YOLOLabel> LabelList { get; set; } = new List<YOLOLabel>() { new YOLOLabel { ID = 1, Name = "person" }, new YOLOLabel { ID = 2, Name = "bicycle" }, new YOLOLabel { ID = 3, Name = "car" }, new YOLOLabel { ID = 4, Name = "motorcycle" }, new YOLOLabel { ID = 5, Name = "airplane" }, new YOLOLabel { ID = 6, Name = "bus" }, new YOLOLabel { ID = 7, Name = "train" }, new YOLOLabel { ID = 8, Name = "truck" }, new YOLOLabel { ID = 9, Name = "boat" }, new YOLOLabel { ID = 10, Name = "traffic light" }, new YOLOLabel { ID = 11, Name = "fire hydrant" }, new YOLOLabel { ID = 12, Name = "stop sign" }, new YOLOLabel { ID = 13, Name = "parking meter" }, new YOLOLabel { ID = 14, Name = "bench" }, new YOLOLabel { ID = 15, Name = "bird" }, new YOLOLabel { ID = 16, Name = "cat" }, new YOLOLabel { ID = 17, Name = "dog" }, new YOLOLabel { ID = 18, Name = "horse" }, new YOLOLabel { ID = 19, Name = "sheep" }, new YOLOLabel { ID = 20, Name = "cow" }, new YOLOLabel { ID = 21, Name = "elephant" }, new YOLOLabel { ID = 22, Name = "bear" }, new YOLOLabel { ID = 23, Name = "zebra" }, new YOLOLabel { ID = 24, Name = "giraffe" }, new YOLOLabel { ID = 25, Name = "backpack" }, new YOLOLabel { ID = 26, Name = "umbrella" }, new YOLOLabel { ID = 27, Name = "handbag" }, new YOLOLabel { ID = 28, Name = "tie" }, new YOLOLabel { ID = 29, Name = "suitcase" }, new YOLOLabel { ID = 30, Name = "frisbee" }, new YOLOLabel { ID = 31, Name = "skis" }, new YOLOLabel { ID = 32, Name = "snowboard" }, new YOLOLabel { ID = 33, Name = "sports ball" }, new YOLOLabel { ID = 34, Name = "kite" }, new YOLOLabel { ID = 35, Name = "baseball bat" }, new YOLOLabel { ID = 36, Name = "baseball glove" }, new YOLOLabel { ID = 37, Name = "skateboard" }, new YOLOLabel { ID = 38, Name = "surfboard" }, new YOLOLabel { ID = 39, Name = "tennis racket" }, new YOLOLabel { ID = 40, Name = "bottle" }, new YOLOLabel { ID = 41, Name = "wine glass" }, new YOLOLabel { ID = 42, Name = "cup" }, new YOLOLabel { ID = 43, Name = "fork" }, new YOLOLabel { ID = 44, Name = "knife" }, new YOLOLabel { ID = 45, Name = "spoon" }, new YOLOLabel { ID = 46, Name = "bowl" }, new YOLOLabel { ID = 47, Name = "banana" }, new YOLOLabel { ID = 48, Name = "apple" }, new YOLOLabel { ID = 49, Name = "sandwich" }, new YOLOLabel { ID = 50, Name = "orange" }, new YOLOLabel { ID = 51, Name = "broccoli" }, new YOLOLabel { ID = 52, Name = "carrot" }, new YOLOLabel { ID = 53, Name = "hot dog" }, new YOLOLabel { ID = 54, Name = "pizza" }, new YOLOLabel { ID = 55, Name = "donut" }, new YOLOLabel { ID = 56, Name = "cake" }, new YOLOLabel { ID = 57, Name = "chair" }, new YOLOLabel { ID = 58, Name = "couch" }, new YOLOLabel { ID = 59, Name = "potted plant" }, new YOLOLabel { ID = 60, Name = "bed" }, new YOLOLabel { ID = 61, Name = "dining table" }, new YOLOLabel { ID = 62, Name = "toilet" }, new YOLOLabel { ID = 63, Name = "tv" }, new YOLOLabel { ID = 64, Name = "laptop" }, new YOLOLabel { ID = 65, Name = "mouse" }, new YOLOLabel { ID = 66, Name = "remote" }, new YOLOLabel { ID = 67, Name = "keyboard" }, new YOLOLabel { ID = 68, Name = "cell phone" }, new YOLOLabel { ID = 69, Name = "microwave" }, new YOLOLabel { ID = 70, Name = "oven" }, new YOLOLabel { ID = 71, Name = "toaster" }, new YOLOLabel { ID = 72, Name = "sink" }, new YOLOLabel { ID = 73, Name = "refrigerator" }, new YOLOLabel { ID = 74, Name = "book" }, new YOLOLabel { ID = 75, Name = "clock" }, new YOLOLabel { ID = 76, Name = "vase" }, new YOLOLabel { ID = 77, Name = "scissors" }, new YOLOLabel { ID = 78, Name = "teddy bear" }, new YOLOLabel { ID = 79, Name = "hair drier" }, new YOLOLabel { ID = 80, Name = "toothbrush" } }; #endregion #region 탐지 사용 여부 - UseDetect /// <summary> /// 탐지 사용 여부 /// </summary> public override bool UseDetect { get; set; } = true; #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Constructor ////////////////////////////////////////////////////////////////////////////////////////// Public #region 생성자 - YOLOCOCOP5Model() /// <summary> /// 생성자 /// </summary> public YOLOCOCOP5Model() { } #endregion } } |
▶ YOLOCOCOP6Model.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 |
using System.Collections.Generic; namespace TestProject { /// <summary> /// YOLO COCO P6 모델 /// </summary> public class YOLOCOCOP6Model : YOLOModel { //////////////////////////////////////////////////////////////////////////////////////////////////// Property ////////////////////////////////////////////////////////////////////////////////////////// Public #region 너비 - Width /// <summary> /// 너비 /// </summary> public override int Width { get; set; } = 1280; #endregion #region 높이 - Height /// <summary> /// 높이 /// </summary> public override int Height { get; set; } = 1280; #endregion #region 깊이 - Depth /// <summary> /// 깊이 /// </summary> public override int Depth { get; set; } = 3; #endregion #region 차원 - Dimension /// <summary> /// 차원 /// </summary> public override int Dimension { get; set; } = 85; #endregion #region 스트라이드 배열 - StrideArray /// <summary> /// 스트라이드 배열 /// </summary> public override float[] StrideArray { get; set; } = new float[] { 8, 16, 32, 64 }; #endregion #region 앵커 배열 - AnchorArray /// <summary> /// 앵커 배열 /// </summary> public override float[][][] AnchorArray { get; set; } = new float[][][] { new float[][] { new float[] { 019, 027 }, new float[] { 044, 040 }, new float[] { 038, 094 } }, new float[][] { new float[] { 096, 068 }, new float[] { 086, 152 }, new float[] { 180, 137 } }, new float[][] { new float[] { 140, 301 }, new float[] { 303, 264 }, new float[] { 238, 542 } }, new float[][] { new float[] { 436, 615 }, new float[] { 739, 380 }, new float[] { 925, 792 } } }; #endregion #region 도형 배열 - ShapeArray /// <summary> /// 도형 배열 /// </summary> public override int[] ShapeArray { get; set; } = new int[] { 160, 80, 40, 20 }; #endregion #region 신뢰도 - Confidence /// <summary> /// 신뢰도 /// </summary> public override float Confidence { get; set; } = 0.20f; #endregion #region 다중 신뢰도 - MultipleConfidence /// <summary> /// 다중 신뢰도 /// </summary> public override float MultipleConfidence { get; set; } = 0.25f; #endregion #region 오버랩 - Overlap /// <summary> /// 오버랩 /// </summary> public override float Overlap { get; set; } = 0.45f; #endregion #region 출력 배열 - OutputArray /// <summary> /// 출력 배열 /// </summary> public override string[] OutputArray { get; set; } = new[] { "output" }; #endregion #region 레이블 리스트 - LabelList /// <summary> /// 레이블 리스트 /// </summary> public override List<YOLOLabel> LabelList { get; set; } = new List<YOLOLabel>() { new YOLOLabel { ID = 1, Name = "person" }, new YOLOLabel { ID = 2, Name = "bicycle" }, new YOLOLabel { ID = 3, Name = "car" }, new YOLOLabel { ID = 4, Name = "motorcycle" }, new YOLOLabel { ID = 5, Name = "airplane" }, new YOLOLabel { ID = 6, Name = "bus" }, new YOLOLabel { ID = 7, Name = "train" }, new YOLOLabel { ID = 8, Name = "truck" }, new YOLOLabel { ID = 9, Name = "boat" }, new YOLOLabel { ID = 10, Name = "traffic light" }, new YOLOLabel { ID = 11, Name = "fire hydrant" }, new YOLOLabel { ID = 12, Name = "stop sign" }, new YOLOLabel { ID = 13, Name = "parking meter" }, new YOLOLabel { ID = 14, Name = "bench" }, new YOLOLabel { ID = 15, Name = "bird" }, new YOLOLabel { ID = 16, Name = "cat" }, new YOLOLabel { ID = 17, Name = "dog" }, new YOLOLabel { ID = 18, Name = "horse" }, new YOLOLabel { ID = 19, Name = "sheep" }, new YOLOLabel { ID = 20, Name = "cow" }, new YOLOLabel { ID = 21, Name = "elephant" }, new YOLOLabel { ID = 22, Name = "bear" }, new YOLOLabel { ID = 23, Name = "zebra" }, new YOLOLabel { ID = 24, Name = "giraffe" }, new YOLOLabel { ID = 25, Name = "backpack" }, new YOLOLabel { ID = 26, Name = "umbrella" }, new YOLOLabel { ID = 27, Name = "handbag" }, new YOLOLabel { ID = 28, Name = "tie" }, new YOLOLabel { ID = 29, Name = "suitcase" }, new YOLOLabel { ID = 30, Name = "frisbee" }, new YOLOLabel { ID = 31, Name = "skis" }, new YOLOLabel { ID = 32, Name = "snowboard" }, new YOLOLabel { ID = 33, Name = "sports ball" }, new YOLOLabel { ID = 34, Name = "kite" }, new YOLOLabel { ID = 35, Name = "baseball bat" }, new YOLOLabel { ID = 36, Name = "baseball glove" }, new YOLOLabel { ID = 37, Name = "skateboard" }, new YOLOLabel { ID = 38, Name = "surfboard" }, new YOLOLabel { ID = 39, Name = "tennis racket" }, new YOLOLabel { ID = 40, Name = "bottle" }, new YOLOLabel { ID = 41, Name = "wine glass" }, new YOLOLabel { ID = 42, Name = "cup" }, new YOLOLabel { ID = 43, Name = "fork" }, new YOLOLabel { ID = 44, Name = "knife" }, new YOLOLabel { ID = 45, Name = "spoon" }, new YOLOLabel { ID = 46, Name = "bowl" }, new YOLOLabel { ID = 47, Name = "banana" }, new YOLOLabel { ID = 48, Name = "apple" }, new YOLOLabel { ID = 49, Name = "sandwich" }, new YOLOLabel { ID = 50, Name = "orange" }, new YOLOLabel { ID = 51, Name = "broccoli" }, new YOLOLabel { ID = 52, Name = "carrot" }, new YOLOLabel { ID = 53, Name = "hot dog" }, new YOLOLabel { ID = 54, Name = "pizza" }, new YOLOLabel { ID = 55, Name = "donut" }, new YOLOLabel { ID = 56, Name = "cake" }, new YOLOLabel { ID = 57, Name = "chair" }, new YOLOLabel { ID = 58, Name = "couch" }, new YOLOLabel { ID = 59, Name = "potted plant" }, new YOLOLabel { ID = 60, Name = "bed" }, new YOLOLabel { ID = 61, Name = "dining table" }, new YOLOLabel { ID = 62, Name = "toilet" }, new YOLOLabel { ID = 63, Name = "tv" }, new YOLOLabel { ID = 64, Name = "laptop" }, new YOLOLabel { ID = 65, Name = "mouse" }, new YOLOLabel { ID = 66, Name = "remote" }, new YOLOLabel { ID = 67, Name = "keyboard" }, new YOLOLabel { ID = 68, Name = "cell phone" }, new YOLOLabel { ID = 69, Name = "microwave" }, new YOLOLabel { ID = 70, Name = "oven" }, new YOLOLabel { ID = 71, Name = "toaster" }, new YOLOLabel { ID = 72, Name = "sink" }, new YOLOLabel { ID = 73, Name = "refrigerator" }, new YOLOLabel { ID = 74, Name = "book" }, new YOLOLabel { ID = 75, Name = "clock" }, new YOLOLabel { ID = 76, Name = "vase" }, new YOLOLabel { ID = 77, Name = "scissors" }, new YOLOLabel { ID = 78, Name = "teddy bear" }, new YOLOLabel { ID = 79, Name = "hair drier" }, new YOLOLabel { ID = 80, Name = "toothbrush" } }; #endregion #region 탐지 사용 여부 - UseDetect /// <summary> /// 탐지 사용 여부 /// </summary> public override bool UseDetect { get; set; } = true; #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Constructor ////////////////////////////////////////////////////////////////////////////////////////// Public #region 생성자 - YOLOCOCOP6Model() /// <summary> /// 생성자 /// </summary> public YOLOCOCOP6Model() { } #endregion } } |
▶ YOLOLabelKind.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 |
namespace TestProject { /// <summary> /// YOLO 레이블 종류 /// </summary> public enum YOLOLabelKind { /// <summary> /// 일반 /// </summary> Generic } } |
▶ YOLOLabel.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 |
using System.Drawing; namespace TestProject { /// <summary> /// YOLO 레이블 /// </summary> public class YOLOLabel { //////////////////////////////////////////////////////////////////////////////////////////////////// Property ////////////////////////////////////////////////////////////////////////////////////////// Public #region ID - ID /// <summary> /// ID /// </summary> public int ID { get; set; } #endregion #region 명칭 - Name /// <summary> /// 명칭 /// </summary> public string Name { get; set; } #endregion #region 종류 - Kind /// <summary> /// YOLO 레이블 종류 /// </summary> public YOLOLabelKind Kind { get; set; } #endregion #region 색상 - Color /// <summary> /// 색상 /// </summary> public Color Color { get; set; } #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Constructor ////////////////////////////////////////////////////////////////////////////////////////// Public #region 생성자 - YOLOLabel() /// <summary> /// 생성자 /// </summary> public YOLOLabel() { Color = Color.Yellow; } #endregion } } |
▶ YOLOPrediction.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 |
using System.Drawing; namespace TestProject { /// <summary> /// YOLO 예측 /// </summary> public class YOLOPrediction { //////////////////////////////////////////////////////////////////////////////////////////////////// Property ////////////////////////////////////////////////////////////////////////////////////////// Public #region 레이블 - Label /// <summary> /// 레이블 /// </summary> public YOLOLabel Label { get; set; } #endregion #region 사각형 - Rectangle /// <summary> /// 사각형 /// </summary> public RectangleF Rectangle { get; set; } #endregion #region 점수 - Score /// <summary> /// 점수 /// </summary> public float Score { get; set; } #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Constructor ////////////////////////////////////////////////////////////////////////////////////////// Public #region 생성자 - YOLOPrediction() /// <summary> /// 생성자 /// </summary> public YOLOPrediction() { } #endregion #region 생성자 - YOLOPrediction(label) /// <summary> /// 생성자 /// </summary> /// <param name="label">레이블</param> public YOLOPrediction(YOLOLabel label) { Label = label; } #endregion #region 생성자 - YOLOPrediction(label, confidence) /// <summary> /// 생성자 /// </summary> /// <param name="label">레이블</param> /// <param name="confidence">신뢰도</param> public YOLOPrediction(YOLOLabel label, float confidence) : this(label) { Score = confidence; } #endregion } } |
▶ YOLOScorer.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 |
using Microsoft.ML.OnnxRuntime; using Microsoft.ML.OnnxRuntime.Tensors; using System; using System.Collections.Generic; using System.Drawing; using System.Drawing.Imaging; using System.IO; using System.Linq; namespace TestProject { /// <summary> /// YOLO 채점기 /// </summary> public class YOLOScorer<T> : IDisposable where T : YOLOModel { //////////////////////////////////////////////////////////////////////////////////////////////////// Field ////////////////////////////////////////////////////////////////////////////////////////// Private #region Field /// <summary> /// 모델 /// </summary> private readonly T model; /// <summary> /// 추론 세션 /// </summary> private readonly InferenceSession inferenceSession; #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Constructor ////////////////////////////////////////////////////////////////////////////////////////// Public #region 생성자 - YOLOScorer() /// <summary> /// 생성자 /// </summary> public YOLOScorer() { this.model = Activator.CreateInstance<T>(); } #endregion #region 생성자 - YOLOScorer(filePath, options) /// <summary> /// 생성자 /// </summary> /// <param name="filePath">파일 경로</param> /// <param name="options">옵션</param> public YOLOScorer(string filePath, SessionOptions options = null) : this() { this.inferenceSession = new InferenceSession(File.ReadAllBytes(filePath), options ?? new SessionOptions()); } #endregion #region 생성자 - YOLOScorer(stream, options) /// <summary> /// 생성자 /// </summary> /// <param name="stream">스트림</param> /// <param name="options">옵션</param> public YOLOScorer(Stream stream, SessionOptions options = null) : this() { using(BinaryReader reader = new BinaryReader(stream)) { this.inferenceSession = new InferenceSession(reader.ReadBytes((int)stream.Length), options ?? new SessionOptions()); } } #endregion #region 생성자 - YOLOScorer(weightByteArray, options) /// <summary> /// 생성자 /// </summary> /// <param name="weightByteArray">가중치 바이트 배열</param> /// <param name="options">옵션</param> public YOLOScorer(byte[] weightByteArray, SessionOptions options = null) : this() { this.inferenceSession = new InferenceSession(weightByteArray, options ?? new SessionOptions()); } #endregion //////////////////////////////////////////////////////////////////////////////////////////////////// Method ////////////////////////////////////////////////////////////////////////////////////////// Public #region 예측하기 - Predict(image) /// <summary> /// 예측하기 /// </summary> /// <param name="image">이미지</param> /// <returns>예측하기</returns> public List<YOLOPrediction> Predict(Image image) { return Supress(ParseOutput(Inference(image), image)); } #endregion #region 리소스 해제하기 - Dispose() /// <summary> /// 리소스 해제하기 /// </summary> public void Dispose() { this.inferenceSession.Dispose(); } #endregion ////////////////////////////////////////////////////////////////////////////////////////// Private #region 크기 변경 비트맵 구하기 - GetResizeBitmap(image) /// <summary> /// 크기 변경 비트맵 구하기 /// </summary> /// <param name="image">이미지</param> /// <returns>크기 변경 비트맵</returns> private Bitmap GetResizeBitmap(Image image) { PixelFormat pixelFormat = image.PixelFormat; Bitmap bitmap = new Bitmap(this.model.Width, this.model.Height, pixelFormat); using(Graphics graphics = Graphics.FromImage(bitmap)) { graphics.Clear(Color.FromArgb(0, 0, 0, 0)); var (widthRatio, heightRatio) = (this.model.Width / (float)image.Width, this.model.Height / (float)image.Height); float ratio = Math.Min(widthRatio, heightRatio); var (width, height) = ((int)(image.Width * ratio), (int)(image.Height * ratio)); var (x, y) = ((this.model.Width / 2) - (width / 2), (this.model.Height / 2) - (height / 2)); graphics.DrawImage(image, new Rectangle(x, y, width, height)); } return bitmap; } #endregion #region 픽셀 추출하기 - ExtractPixels(image) /// <summary> /// 픽셀 추출하기 /// </summary> /// <param name="image">이미지</param> /// <returns>텐서</returns> private Tensor<float> ExtractPixels(Image image) { Bitmap bitmap = new Bitmap(image); Rectangle rectangle = new Rectangle(0, 0, image.Width, image.Height); BitmapData bitmapData = bitmap.LockBits(rectangle, ImageLockMode.ReadOnly, image.PixelFormat); DenseTensor<float> tensor = new DenseTensor<float>(new[] { 1, 3, image.Height, image.Width }); unsafe { for(int y = 0; y < bitmapData.Height; y++) { byte* pointer = (byte*)bitmapData.Scan0 + (y * bitmapData.Stride); for(int x = 0; x < bitmapData.Width; x++) { tensor[0, 0, y, x] = pointer[x * 3 + 0] / 255.0f; tensor[0, 1, y, x] = pointer[x * 3 + 1] / 255.0f; tensor[0, 2, y, x] = pointer[x * 3 + 2] / 255.0f; } } bitmap.UnlockBits(bitmapData); } return tensor; } #endregion #region 추론하기 - Inference(image) /// <summary> /// 추론하기 /// </summary> /// <param name="image">이미지</param> /// <returns>밀집 텐서 배열</returns> private DenseTensor<float>[] Inference(Image image) { Bitmap resizedBitmap = null; if(image.Width != this.model.Width || image.Height != this.model.Height) { resizedBitmap = GetResizeBitmap(image); } List<NamedOnnxValue> inputList = new List<NamedOnnxValue> { NamedOnnxValue.CreateFromTensor("images", ExtractPixels(resizedBitmap ?? image)) }; IDisposableReadOnlyCollection<DisposableNamedOnnxValue> resultCollection = this.inferenceSession.Run(inputList); List<DenseTensor<float>> outputList = new List<DenseTensor<float>>(); foreach(string output in this.model.OutputArray) { outputList.Add(resultCollection.First(x => x.Name == output).Value as DenseTensor<float>); }; return outputList.ToArray(); } #endregion #region 탐지 파싱하기 - ParseDetect(outputTensor, image) /// <summary> /// 탐지 파싱하기 /// </summary> private List<YOLOPrediction> ParseDetect(DenseTensor<float> outputTensor, Image image) { List<YOLOPrediction> resultList = new List<YOLOPrediction>(); var (xGain, yGain) = (this.model.Width / (float)image.Width, this.model.Height / (float)image.Height); float gain = Math.Min(xGain, yGain); var (xPadding, yPadding) = ((this.model.Width - image.Width * gain) / 2, (this.model.Height - image.Height * gain) / 2); for(int i = 0; i < outputTensor.Length / this.model.Dimension; i++) { if(outputTensor[0, i, 4] <= this.model.Confidence) { continue; } for(int j = 5; j < this.model.Dimension; j++) { outputTensor[0, i, j] = outputTensor[0, i, j] * outputTensor[0, i, 4]; } for(int k = 5; k < this.model.Dimension; k++) { if(outputTensor[0, i, k] <= this.model.MultipleConfidence) { continue; } float xMinimum = ((outputTensor[0, i, 0] - outputTensor[0, i, 2] / 2) - xPadding) / gain; float yMinimum = ((outputTensor[0, i, 1] - outputTensor[0, i, 3] / 2) - yPadding) / gain; float xMaximum = ((outputTensor[0, i, 0] + outputTensor[0, i, 2] / 2) - xPadding) / gain; float yMaximum = ((outputTensor[0, i, 1] + outputTensor[0, i, 3] / 2) - yPadding) / gain; xMinimum = Clamp(xMinimum, 0, image.Width ); yMinimum = Clamp(yMinimum, 0, image.Height); xMaximum = Clamp(xMaximum, 0, image.Width ); yMaximum = Clamp(yMaximum, 0, image.Height); YOLOLabel label = this.model.LabelList[k - 5]; YOLOPrediction prediction = new YOLOPrediction(label, outputTensor[0, i, k]) { Rectangle = new RectangleF(xMinimum, yMinimum, xMaximum - xMinimum, yMaximum - yMinimum) }; resultList.Add(prediction); } } return resultList; } #endregion #region XYXY 배열 구하기 - GetXYXYArray(sourceArray) /// <summary> /// XYXY 배열 구하기 /// </summary> /// <param name="sourceArray">소스 배열</param> /// <returns>XYXY 배열</returns> /// <remarks>xywh bbox 형식을 xyxy로 변환한다.</remarks> private float[] GetXYXYArray(float[] sourceArray) { float[] targetArray = new float[4]; targetArray[0] = sourceArray[0] - sourceArray[2] / 2f; targetArray[1] = sourceArray[1] - sourceArray[3] / 2f; targetArray[2] = sourceArray[0] + sourceArray[2] / 2f; targetArray[3] = sourceArray[1] + sourceArray[3] / 2f; return targetArray; } #endregion #region 범위 내 자르기 - Clamp(value, minimumValue, maximumValue) /// <summary> /// 범위 내 자르기 /// </summary> /// <param name="value">값</param> /// <param name="minimumValue">최소값</param> /// <param name="maximumValue">최대값</param> /// <returns>값</returns> public float Clamp(float value, float minimumValue, float maximumValue) { return (value < minimumValue) ? minimumValue : (value > maximumValue) ? maximumValue : value; } #endregion #region 시그모이드 구하기 - GetSigmoid(value) /// <summary> /// 시그모이드 구하기 /// </summary> /// <param name="value">값</param> /// <returns>시그모이드</returns> private float GetSigmoid(float value) { return 1 / (1 + (float)Math.Exp(-value)); } #endregion #region 시그모이드 파싱하기 - ParseSigmoid(outputTensor, image) /// <summary> /// 시그모이드 파싱하기 /// </summary> /// <param name="outputTensor">출력 텐서</param> /// <param name="image">이미지</param> /// <returns>예측 리스트</returns> private List<YOLOPrediction> ParseSigmoid(DenseTensor<float>[] outputTensor, Image image) { List<YOLOPrediction> resultList = new List<YOLOPrediction>(); var (xGain, yGain) = (this.model.Width / (float)image.Width, this.model.Height / (float)image.Height); float gain = Math.Min(xGain, yGain); var (xPadding, yPadding) = ((this.model.Width - image.Width * gain) / 2, (this.model.Height - image.Height * gain) / 2); for(int i = 0; i < outputTensor.Length; i++) { int shapeCount = this.model.ShapeArray[i]; for(int j = 0; j < this.model.AnchorArray.Length; j++) { for(int y = 0; y < shapeCount; y++) { for(int x = 0; x < shapeCount; x++) { int offset = (shapeCount * shapeCount * j + shapeCount * y + x) * this.model.Dimension; float[] bufferArray = outputTensor[i].Skip(offset).Take(this.model.Dimension).Select(GetSigmoid).ToArray(); float objectConfidence = bufferArray[4]; if(objectConfidence < this.model.Confidence) { continue; } List<float> scoreList = bufferArray.Skip(5).Select(b => b * objectConfidence).ToList(); float multipleConfidence = scoreList.Max(); if(multipleConfidence <= this.model.MultipleConfidence) { continue; } float rawX = (bufferArray[0] * 2 - 0.5f + x) * this.model.StrideArray[i]; float rawY = (bufferArray[1] * 2 - 0.5f + y) * this.model.StrideArray[i]; float rawWidth = (float)Math.Pow(bufferArray[2] * 2, 2) * this.model.AnchorArray[i][j][0]; float rawHeight = (float)Math.Pow(bufferArray[3] * 2, 2) * this.model.AnchorArray[i][j][1]; float[] xyxyArray = GetXYXYArray(new float[] { rawX, rawY, rawWidth, rawHeight }); float xMinimum = Clamp((xyxyArray[0] - xPadding) / gain, 0, image.Width ); float yMinimum = Clamp((xyxyArray[1] - yPadding) / gain, 0, image.Height); float xMaximum = Clamp((xyxyArray[2] - xPadding) / gain, 0, image.Width ); float yMaximum = Clamp((xyxyArray[3] - yPadding) / gain, 0, image.Height); YOLOLabel label = this.model.LabelList[scoreList.IndexOf(multipleConfidence)]; YOLOPrediction prediction = new YOLOPrediction(label, multipleConfidence) { Rectangle = new RectangleF(xMinimum, yMinimum, xMaximum - xMinimum, yMaximum - yMinimum) }; resultList.Add(prediction); } } } } return resultList; } #endregion #region 출력 파싱하기 - ParseOutput(outputTenserArray, image) /// <summary> /// 출력 파싱하기 /// </summary> /// <param name="outputTenserArray">출력 텐서 배열</param> /// <param name="image">이미지</param> /// <returns>예측 리스트</returns> private List<YOLOPrediction> ParseOutput(DenseTensor<float>[] outputTenserArray, Image image) { return this.model.UseDetect ? ParseDetect(outputTenserArray[0], image) : ParseSigmoid(outputTenserArray, image); } #endregion #region 억제하기 - Supress(sourceList) /// <summary> /// 억제하기 /// </summary> /// <param name="sourceList">소스 리스트</param> /// <returns>억제 리스트</returns> private List<YOLOPrediction> Supress(List<YOLOPrediction> sourceList) { List<YOLOPrediction> targetList = new List<YOLOPrediction>(sourceList); foreach(YOLOPrediction source in sourceList) { foreach(YOLOPrediction target in targetList.ToList()) { if(target == source) { continue; } var (rectangle1, rectangle2) = (source.Rectangle, target.Rectangle); RectangleF intersectionRectangle = RectangleF.Intersect(rectangle1, rectangle2); float intesectionArea = intersectionRectangle.GetArea(); float unionArea = rectangle1.GetArea() + rectangle2.GetArea() - intesectionArea; float overlap = intesectionArea / unionArea; if(overlap > this.model.Overlap) { if(source.Score > target.Score) { targetList.Remove(target); } } } } return targetList; } #endregion } } |
▶ Program.cs
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 |
using System; using System.Collections.Generic; using System.Drawing; namespace TestProject { /// <summary> /// 프로그램 /// </summary> class Program { //////////////////////////////////////////////////////////////////////////////////////////////////// Method ////////////////////////////////////////////////////////////////////////////////////////// Static //////////////////////////////////////////////////////////////////////////////// Private #region 프로그램 시작하기 - Main() /// <summary> /// 프로그램 시작하기 /// </summary> private static void Main() { using Image image = Image.FromFile("IMAGE//source.jpg"); //using YOLOScorer<YOLOCOCOP5Model> scorer = new YOLOScorer<YOLOCOCOP5Model>("ASSET/yolov5s.onnx"); using YOLOScorer<YOLOCOCOP6Model> scorer = new YOLOScorer<YOLOCOCOP6Model>("ASSET/yolov5s6.onnx"); List<YOLOPrediction> predictionList = scorer.Predict(image); using Graphics graphics = Graphics.FromImage(image); foreach(YOLOPrediction prediction in predictionList) { double score = Math.Round(prediction.Score, 2); graphics.DrawRectangles(new Pen(prediction.Label.Color, 1), new[] { prediction.Rectangle }); var (x, y) = (prediction.Rectangle.X - 3, prediction.Rectangle.Y - 23); graphics.DrawString ( $"{prediction.Label.Name} ({score})", new Font("Consolas", 16, GraphicsUnit.Pixel), new SolidBrush(prediction.Label.Color), new PointF(x, y) ); } image.Save("d:\\result.jpg"); } #endregion } } |
