detección de caídas de python, detección de caídas detección de comportamiento de posición abierta

detección de caídas de python, detección de caídas detección de comportamiento de posición abierta

import cv2
import numpy as np
from torch import from_numpy, jit
from modules.keypoints import extract_keypoints, group_keypoints
from modules.pose import Pose
from action_detect.detect import action_detect
import os
from math import ceil, floor


os.environ["PYTORCH_JIT"] = "0"

class ImageReader(object):
    def __init__(self, file_names):
        self.file_names = file_names
        self.max_idx = len(file_names)

    def __iter__(self):
        self.idx = 0
        return self

    def __next__(self):
        if self.idx == self.max_idx:
            raise StopIteration
        img = cv2.imread(self.file_names[self.idx], cv2.IMREAD_COLOR)
        if img.size == 0:
            raise IOError('Image {} cannot be read'.format(self.file_names[self.idx]))
        self.idx = self.idx + 1
        return img


class VideoReader(object):
    def __init__(self, file_name,code_name):
        self.file_name = file_name
        self.code_name = str(code_name)
        try:  # OpenCV needs int to read from webcam
            self.file_name = int(file_name)
        except ValueError:
            pass

    def __iter__(self):
        self.cap = cv2.VideoCapture(self.file_name)#读入已有视频检测
        #self.cap = cv2.VideoCapture(0)#调用笔记本内置摄像头检测
        if not self.cap.isOpened():
            raise IOError('Video {} cannot be opened'.format(self.file_name))
        return self

    def __next__(self):
        was_read, img = self.cap.read()
        if not was_read:
            raise StopIteration

        # print(self.cap.get(7),self.cap.get(5))
        cv2.putText(img,self.code_name, (5,35),
                                cv2.FONT_HERSHEY_COMPLEX, 1, (0, 0, 255))
        return img

def normalize(img, img_mean, img_scale):
    img = np.array(img, dtype=np.float32)
    img = (img - img_mean) * img_scale
    return img


def pad_width(img, stride, pad_value, min_dims):
    h, w, _ = img.shape
    h = min(min_dims[0], h)
    min_dims[0] = ceil(min_dims[0] / float(stride)) * stride
    min_dims[1] = max(min_dims[1], w)
    min_dims[1] = ceil(min_dims[1] / float(stride)) * stride
    pad = []
    pad.append(int(floor((min_dims[0] - h) / 2.0)))
    pad.append(int(floor((min_dims[1] - w) / 2.0)))
    pad.append(int(min_dims[0] - h - pad[0]))
    pad.append(int(min_dims[1] - w - pad[1]))
    padded_img = cv2.copyMakeBorder(img, pad[0], pad[2], pad[1], pad[3],
                                    cv2.BORDER_CONSTANT, value=pad_value)
    return padded_img, pad

def infer_fast(net, img, net_input_height_size, stride, upsample_ratio, cpu,
               pad_value=(0, 0, 0), img_mean=(128, 128, 128), img_scale=1/256):
    height, width, _ = img.shape
    scale = net_input_height_size / height

    scaled_img = cv2.resize(img, (0, 0), fx=scale, fy=scale, interpolation=cv2.INTER_CUBIC)
    scaled_img = normalize(scaled_img, img_mean, img_scale)
    min_dims = [net_input_height_size, max(scaled_img.shape[1], net_input_height_size)]
    padded_img, pad = pad_width(scaled_img, stride, pad_value, min_dims)

    tensor_img = from_numpy(padded_img).permute(2, 0, 1).unsqueeze(0).float()
    if not cpu:
        #tensor_img = tensor_img.cuda()
        pass

    stages_output = net(tensor_img)

    # print(stages_output)

    stage2_heatmaps = stages_output[-2]
    heatmaps = np.transpose(stage2_heatmaps.squeeze().cpu().data.numpy(), (1, 2, 0))
    heatmaps = cv2.resize(heatmaps, (0, 0), fx=upsample_ratio, fy=upsample_ratio, interpolation=cv2.INTER_CUBIC)

    stage2_pafs = stages_output[-1]
    pafs = np.transpose(stage2_pafs.squeeze().cpu().data.numpy(), (1, 2, 0))
    pafs = cv2.resize(pafs, (0, 0), fx=upsample_ratio, fy=upsample_ratio, interpolation=cv2.INTER_CUBIC)

    return heatmaps, pafs, scale, pad

Python puede generar un video para llamar a la cámara basado en la detección de caídas de posición abierta: se busca programador

https://download.csdn.net/download/babyai996/85075816