DL之Mask R-CNN：2018.6.26世界杯阿根廷队VS尼日利亚比赛2:1实现Mask R-CNN目标检测

DL之Mask R-CNN：2018.6.26世界杯阿根廷队VS尼日利亚比赛2:1实现Mask R-CNN目标检测

目录

输出结果

人身检测

核心代码

输出结果

先上目标检测结果

人身检测

观众席：众里寻他千百度，蓦然回首，那人却在，灯火阑珊处。藏得再深也能找到你……

核心代码

import os
import sys
import random
import math
import numpy as np
import skimage.io
import matplotlib
import matplotlib.pyplot as plt
 
 Root directory of the project
ROOT_DIR = os.path.abspath("../")
 
 Import Mask RCNN
sys.path.append(ROOT_DIR)   To find local version of the library
from mrcnn import utils
import mrcnn.model as modellib
from mrcnn import visualize
 Import COCO config
sys.path.append(os.path.join(ROOT_DIR, "samples/coco/"))   To find local version
import coco
 
 Directory to save logs and trained model
MODEL_DIR = os.path.join(ROOT_DIR, "logs")
 
 Local path to trained weights file
COCO_MODEL_PATH = os.path.join(ROOT_DIR, "mask_rcnn_coco.h5")
 Download COCO trained weights from Releases if needed
if not os.path.exists(COCO_MODEL_PATH):
    utils.download_trained_weights(COCO_MODEL_PATH)
 
 Directory of images to run detection on
IMAGE_DIR = os.path.join(ROOT_DIR, "images01")
 
 
class InferenceConfig(coco.CocoConfig):
     Set batch size to 1 since we'll be running inference on
     one image at a time. Batch size = GPU_COUNT * IMAGES_PER_GPU
    GPU_COUNT = 1
    IMAGES_PER_GPU = 1
 
config = InferenceConfig()
config.display()
 
Create Model and Load Trained Weights
 Create model object in inference mode.
model = modellib.MaskRCNN(mode="inference", model_dir=MODEL_DIR, config=config)
 
 Load weights trained on MS-COCO
model.load_weights(COCO_MODEL_PATH, by_name=True)

Configurations:
BACKBONE                       resnet101
BACKBONE_STRIDES               [4, 8, 16, 32, 64]
BATCH_SIZE                     1
BBOX_STD_DEV                   [0.1 0.1 0.2 0.2]
COMPUTE_BACKBONE_SHAPE         None
DETECTION_MAX_INSTANCES        100
DETECTION_MIN_CONFIDENCE       0.7
DETECTION_NMS_THRESHOLD        0.3
FPN_CLASSIF_FC_LAYERS_SIZE     1024
GPU_COUNT                      1
GRADIENT_CLIP_NORM             5.0
IMAGES_PER_GPU                 1
IMAGE_MAX_DIM                  1024
IMAGE_META_SIZE                93
IMAGE_MIN_DIM                  800
IMAGE_MIN_SCALE                0
IMAGE_RESIZE_MODE              square
IMAGE_SHAPE                    [1024 1024    3]
LEARNING_MOMENTUM              0.9
LEARNING_RATE                  0.001
LOSS_WEIGHTS                   {'rpn_class_loss': 1.0, 'rpn_bbox_loss': 1.0, 'mrcnn_class_loss': 1.0, 'mrcnn_bbox_loss': 1.0, 'mrcnn_mask_loss': 1.0}
-selector-tag">MASK_POOL_SIZE                 14
-selector-tag">MASK_SHAPE                     [28, 28]
-selector-tag">MAX_GT_INSTANCES               100
-selector-tag">MEAN_PIXEL                     [123.7 116.8 103.9]
-selector-tag">MINI_MASK_SHAPE                (56, 56)
-selector-tag">NAME                           -selector-tag">coco
-selector-tag">NUM_CLASSES                    81
-selector-tag">POOL_SIZE                      7
-selector-tag">POST_NMS_ROIS_INFERENCE        1000
-selector-tag">POST_NMS_ROIS_TRAINING         2000
-selector-tag">ROI_POSITIVE_RATIO             0.33
-selector-tag">RPN_ANCHOR_RATIOS              [0.5, 1, 2]
-selector-tag">RPN_ANCHOR_SCALES              (32, 64, 128, 256, 512)
-selector-tag">RPN_ANCHOR_STRIDE              1
-selector-tag">RPN_BBOX_STD_DEV               [0.1 0.1 0.2 0.2]
-selector-tag">RPN_NMS_THRESHOLD              0.7
-selector-tag">RPN_TRAIN_ANCHORS_PER_IMAGE    256
-selector-tag">STEPS_PER_EPOCH                1000
-selector-tag">TOP_DOWN_PYRAMID_SIZE          256
-selector-tag">TRAIN_BN                       -selector-tag">False
-selector-tag">TRAIN_ROIS_PER_IMAGE           200
-selector-tag">USE_MINI_MASK                  -selector-tag">True
-selector-tag">USE_RPN_ROIS                   -selector-tag">True
-selector-tag">VALIDATION_STEPS               50
-selector-tag">WEIGHT_DECAY                   0.0001

Processing 1 images
image                    shape-operator">: -punctuation">(506-punctuation">, 900-punctuation">, 3-punctuation">)         min-operator">:    0.00000  max-operator">:  255.00000  uint8
molded_images            shape-operator">: -punctuation">(1-punctuation">, 1024-punctuation">, 1024-punctuation">, 3-punctuation">)    min-operator">: -operator">-123.70000  max-operator">:  151.10000  float64
image_metas              shape-operator">: -punctuation">(1-punctuation">, 93-punctuation">)               min-operator">:    0.00000  max-operator">: 1024.00000  float64
anchors                  shape-operator">: -punctuation">(1-punctuation">, 261888-punctuation">, 4-punctuation">)        min-operator">:   -operator">-0.35390  max-operator">:    1.29134  float32