保存faster-rcnn的检测结果

最新推荐文章于 2021-11-26 10:22:29 发布

v1_vivian

最新推荐文章于 2021-11-26 10:22:29 发布

阅读量3.1k

点赞数 3

分类专栏：深度学习教程目标检测文章标签：神经网络 faster-rcnn 深度学习机器学习

本文链接：https://blog.csdn.net/v1_vivian/article/details/73864151

版权

本文介绍如何保存Faster R-CNN在测试时的检测结果，以便后续进行分析。内容包括图像名称、类别信息及边界框坐标等关键数据。

摘要由CSDN通过智能技术生成

为了分析faster-Rcnn的测试结果，需要先将测试结果保存起来，效果如下：

（图片名类别 bbox坐标）

代码如下：

#!/usr/bin/env python

# --------------------------------------------------------
# Faster R-CNN
# Copyright (c) 2015 Microsoft
# Licensed under The MIT License [see LICENSE for details]
# Written by Ross Girshick
# --------------------------------------------------------

"""
Demo script showing detections in sample images.

See README.md for installation instructions before running.
"""

import _init_paths
#import matplotlib
#matplotlib.use('Agg')
from fast_rcnn.config import cfg
from fast_rcnn.test import im_detect
from fast_rcnn.nms_wrapper import nms
from utils.timer import Timer
import matplotlib.pyplot as plt
import numpy as np
import scipy.io as sio
import caffe, os, sys, cv2
import argparse


#自己的类别名称
CLASSES = ('__background__',
           'xiansu5', 'xiansu10', 'xiansu15', 'xiansu20',
           'xiansu30', 'xiansu40', 'xiansu50', 'xiansu60', 'xiansu70',
           'xiansu80', 'xiansu90', 'xiansu100','xiansu110', 'xiansu120')

NETS = {'vgg16': ('VGG16',
                  'VGG16_faster_rcnn_final.caffemodel'),
        'zf': ('ZF',
                  'ZF_faster_rcnn_final.caffemodel')}


def vis_detections(image_name, class_name, dets, thresh=0.5):
    """Draw detected bounding boxes."""
    inds = np.where(dets[:, -1] >= thresh)[0]
    if len(inds) == 0:
        return

    for