dasflow 0.0.7

DAS event detection flow based on deep learning

Das event detection flow

A high-dimensional template matching framework based on PyTorch.

Installation

pip install dasflow

git clone

pip only supports the installation of the latest version of the package without the tutorial. If you want to use the tutorial and test data, you can clone the repository and install it locally.

https://gitee.com/gemini_lh/dasflow.git

Refer to the doc.ipynb notebook for a step-by-step tutorial.

Usage

TMF Example

import numpy as np
from dasflow import tmf
# generate a random template
tmp = np.random.rand(10, 100, 20)
# generate a random image
data = np.random.rand(100, 1000)
# calculate the cross-correlation between the template and the image
corr = tmf.tma(data,tmp, step=1,device='cpu',moves = [],is_sum=False,batch_size=-1,half=False,save_memory=False)

Parameters

• tmp (numpy.ndarray or torch.Tensor): The template to be matched.
• data (numpy.ndarray or torch.Tensor): The image to search for the template.
• step (int, optional): The step size of the convolution. Defaults to 1.
• device (str, optional): The device to perform the computation on. Defaults to 'cpu'.
• moves (list, optional): A list of moves to apply to the template before matching. Defaults to [].
• batch_size (int, optional): The batch size to use for the computation. Defaults to -1.
• save_memory (bool, optional): Whether to use half-precision floating point numbers to save memory. Defaults to False.

Returns

• numpy.ndarray: The cross-correlation between the template and the image.

Hough Example

from dasflow import hough
import numpy as np
data = np.random.randn(512, 512)
hough(
    data,
    freq=100,
    bandpass=[2,8],
    sl=[10,20],
    resample=1,
    sigma=1.3,
    low_threshold=3,
    high_threshold=6,
    theta=np.linspace(np.pi/2/90* 10/100,np.pi/2/90 *10,99), # 0~10度
    fil='bandpass',
    S_L=True,
    beta=0,
    kernel=(3,3)
)

Parameters

• data (numpy.ndarray or torch.Tensor): The image to search for the template.
• freq (int, optional): The frequency of the template. Defaults to 100.
• bandpass (list, optional): The bandpass filter to apply to the image. Defaults to [2,8].
• sl (list, optional): The size of the template. Defaults to [10,20].
• resample (int, optional): The resample rate of the image. Defaults to 1.
• sigma (float, optional): The sigma of the Gaussian filter. Defaults to 1.3.
• low_threshold (float, optional): The low threshold of the Canny edge detector. Defaults to 3.
• high_threshold (float, optional): The high threshold of the Canny edge detector. Defaults to 6.
• theta (numpy.ndarray, optional): The theta of the Hough transform. Defaults to np.linspace(np.pi/2/9010/100,np.pi/2/9010,99).
• fil (str, optional): The filter to apply to the image. Defaults to 'bandpass'.
• S_L (bool, optional): Whether to apply the Laplacian filter to the image. Defaults to True.
• beta (float, optional): The beta of the Laplacian filter. Defaults to 0.
• kernel (tuple, optional): The kernel size of the Laplacian filter. Defaults to (3,3).

ADE-mini Example

from dasflow.model import Mini
model = Mini()
data = torch.randn(1, 1, 512, 512)
model(data)

ADE detector Example

from dasflow.model import Mini
from matplotlib import pyplot as plt
from dasflow.model_detect import detector
import torch
model = Mini()
model.load_state_dict(torch.load('dasflow/model_all.pth',map_location=torch.device('cpu')))
detector = detector(model,'cpu')
# 使用detector进行检测
ans = detector.detect(
    data, 
    fil='bandpass', # 滤波器类型，可选值为'bandpass'或'Gauss'，默认为'bandpass'
    S_L=True, # 是否应用sta/lta算法，可选值为True或False，默认为True
    bandpass=[2,8], # bandpass滤波器的频率范围，可选值为[low,high]，默认为[2,8]
    freq=200,  # 采样频率，默认为100
    sl=[0.1,1], # sta/lta算法的sta和lta时间窗口，可选值为[sta,lta]，默认为[0.5,1]
    beta=0, # 高斯滤波器的标准差，默认为0
    kernel=(3,3), # 高斯滤波器的核大小，默认为(3,3)
    hwin=512,  # 时间窗口大小，默认为512
    wwin=512,  # 道窗口大小，默认为512
    overlap=0.25, # 重叠率，默认为0.25
    figure=True # 是否显示检测结果，默认为False
    )

Parameters

• data (torch.Tensor): The input data to the model.

License

MIT License

Copyright (c) [2023] []

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

References