Sogou PhotoShop image processing algorithm summary

Not regularly updated...... Layer hybrid algorithm:   One of the PS layer algorithm (opacity, multiply, color depth, Color Dodge) Image adjustment algorithm PS filter algorithm...
Read(1504) comment(0)

Machine vision of oriented gradients Histogram

Of oriented gradients Histogram HoG is a very important feature in the field of computer vision and image processing. It is widely used in object detection, face detection, face detection and so on. HoG was originally proposed by Dalal Navneet and Triggs Bill on the CVPR in 2005. HoG algorithm is very simple, the characteristics of the object is very effective. Simple and efficient, which is probably also from being raised, it is widely used by the CV sector...
Read(458) comment(0)

Scikit-image canny Python: edge detection

This use case illustrates the use of Canny edge detection Numpy as NP import Matplotlib.pyplot as PLT import SciPy import ndimage as NDI from Skimage import feature from Generate noisy image of a square # Im = np.zeros ((128, 128)) Im[3...
Read(290) comment(0)

Machine vision Binary Pattern Local (LBP)

Binary pattern Local (LBP), in the field of machine vision, is a very important feature. LBP can effectively deal with the change of illumination, and it is widely used in texture analysis and texture recognition. LBP algorithm is very simple, in simple terms is the gray of a pixel in the image value and its neighborhood pixel gray value, as shown below: if the values of neighboring pixels than the big, assigned to 1, on the contrary, it is assigned to 0, begin from the upper left corner like can form a bit of a chain, then the bit chain...
Read(376) comment(0)

Machine learning F1-Score, recall, precision

In machine learning, pattern recognition, we do classification, will use some indicators to judge the pros and cons of the algorithm, the most commonly used is the recognition rate, simply speaking, is Acc=Npre/Ntotalacc=N_{pre}/N_{total} The NpreN_{pre} here represents the number of samples predicted on the NtotalN_{total}, representing the total number of samples tested. Recognition rate is too simple, can not fully reflect the performance of the algorithm, in addition to recognition rate, there are a number of commonly used indicators, that we have to introduce...
Read(455) comment(0)

Machine vision Feature Gabor

In machine vision, Gabor feature is a more common feature, because it can be very good to simulate human visual impact response and is widely used in image processing, Gabor feature is generally through the image with Gabor filter convolution and get, Gabor filter is defined as a product of a Gaussian function with sine function, the expressions are as follows: G (x, y; lambda, theta, psi, sigma, gamma (=exp) - x '2+ of 2Y' 22 sigma gamma 2 (EXP) I (x '+ 2 pi lambda psi)) G (x,...
Read(592) comment(1)

The basic operation of scikit-image Python: image

This use case illustrates the basic operation of the Python image. Numpy as NP import Skimage import data from Matplotlib.pyplot as PLT import Camera = data.camera () # image in front of the 10 elements of a value of 0 Camera[: 10] = 0 # for image pixel values of pixels is less than 87 Mask = camera 87 # will find the point of fu...
Read(540) comment(0)

Scikit-image binary descriptor Python:

This use case illustrates binary description algorithm BRIEF Skimage import data from Skimage import transform as TF from Skimage.feature import (match_descriptors, corner_peaks, corner_harris, from,...
Read(502) comment(0)

Scikit-image Blob detection Python:

This case mainly introduces containing image blob detection using three kinds of algorithms, or blob called spots, is in a picture, dark background and bright region, or bright background of dark areas, can be called a blob. The main use of the contrast between blob and background to detect. Three algorithms are introduced for this use case; Of Gaussian Laplacian (LoG) This is the slowest, but the most accurate algorithm, simple, that is, a map of a series of different size of the Gauss filter, and then the filtered image to do...
Read(923) comment(0)

Scikit-image Python: color image filtering

Filter are in gray image, scikit-image provides for color image filtering Decorator:adapt_rgb RGB, an RGB provides two forms of filtering, a is the RGB three channels separately. Another way is to convert the RGB to HSV color model, then for the V channel processing, finally reversed RGB color model. For mode one, known as each_channel @adapt_rgb (each_channel) Def...
Read(885) comment(0)

Neural Networks Python:

This is achieved using Neural Networks Python, based on 2.7.9 numpy, Python, matplotlib. Code from the Stanford University curriculum: http://cs231n.github.io/neural-networks-case-study/ Basically copied over, through this program will help to understand the python syntax, as well as the principle of Networks Neural. Numpy a import...
Read(789) comment(2)

PS filter - sparkle effect

Clc; All clear; All close; Addpath (Algortihm\Image Processing\PS Algorithm''E:\PhotoShop); I=imread ('4.jpg'); Image=double (I) /255; [height, width, depth]=size (Image); Rays = 25; Radius = 25; A...
Read(855) comment(0)

PS filter - lens flare

Clc; All clear; All close; Addpath (Algortihm\Image Processing\PS Algorithm''E:\PhotoShop); I=imread ('4.jpg'); Image=double (I) /255*0; [height, width, depth]=size (Image); Rays = 200; Radius=100; B...
Read(855) comment(0)

PS color gradient

Clc; All clear; All close; Addpath (Algortihm\Image Processing\PS Algorithm''E:\PhotoShop); I=imread ('4.jpg'); Image=double (I) /255; [height, width, depth]=size (Image); RNW=1.0; gNW=0.0; bN...
Read(779) comment(0)

PS image adjustment - and bias gain

Clc; All clear; All close; Addpath (Algortihm\Image Processing\PS Algorithm''E:\PhotoShop); Image=imread ('4.jpg'); Image=double (Image) /255; % imshow (Image) %% % the gain value 0-1 set % the set...
Read(760) comment(0)

Soft_max Python: classifier

This is achieved using soft Max Python classifier, based on 2.7.9 numpy, Python, matplotlib. Code from the Stanford University curriculum: http://cs231n.github.io/neural-networks-case-study/ Basically copied over, through this program will help to understand the syntax of python. Numpy as NP import Matplotlib.pyp import...
Read(825) comment(0)
130 data a total of 9 pagesOne Two Three Four Five ... Next page Shadowe
    personal data
    • Visit109368 times
    • Integral:Two thousand four hundred and fifty-eight
    • Grade
    • Rank:8814th name
    • Original101
    • Reproduced:0
    • Translation:29
    • Comments:15
    Personal introduction
    Focus on image processing, computer vision, machine learning. I hope you can write beautiful code and text. Through the trials of a long journey to find possible, in the dimension of the world.
    Blog column