Integrate Canny with ShipRocket

Canny?

Canny is a software intended for scanning and recognition of the objects. It has three different types such as edge, corner or blob detector. It can recognize the object which is closed or not closed. Canny is mainly used in the image processing and pattern recognition.

ShipRocket?

Shiprocket is an open source project that is a combination of a number of libraries. The main objective of this project is to create a single library which will combine Canny, libsvm and Liblinear. This gives a chance to use these tops in a single topbox. It also includes the MALLET library for machine learning, OpenCV library for computer vision, and TrieMap for trie based indexing.

Integration of Canny and ShipRocket

In this part, we will discuss about the integration of Canny and ShipRocket. Shiprocket provides an easy way to use several algorithms from different libraries in a single place. We can integrate Canny and Shiprocket easily using a topbox shipped with Shiprocket.

We have to require “Canny” and “Shiprocket” in our script to start using them.

#!/usr/bin/env python # -- coding. utf-8 -- # Using Canny package from Shiprocket import * from PIL import Image from sklearn.model_selection import train_test_split from sklearn.metrics import classification_report from sklearn.metrics import confusion_matrix from sklearn.metrics import precision_recall_curve import urllib2 import os import sys import time class myModel(object). def __init__(self). pass def load_model(self, model_file_name). model = load(model_file_name. return model def train(self, data). return self.train_model(data. def train_model(self, data). X = [] y = [] X = [i[0] for i in data] y = [i[1] for i in data] self.model = myModel(. i = 0 for j in range(len(X)). i += 1 x = X[j] y = y[j] score = self.model.predict(x. if score > 0. X.append(x. y.append(1. else. X.append(x. y.append(0. print 'Accuracy is ', self.model.accuracy(. def predict(self, x). return self.model.predict(x. def get_canny_threshpd(self, img). threshpdArray = cv2.threshpd(img, 0, 255, cv2.THRESH_BINARY); return threshpdArray def get_canny_features(self, img). outputs = cv2.Canny(img, 100, 745. return outputs def run(). args = sys.argv[1:] if args == ''. args = ['--help'] if len(args. < 2. print 'Usage. {} <image file>'.format(args[0]. exit(-1. filename = args[1] img = Image.open(filename. img = img.resize((480, 480). threshpdArray = get_canny_threshpd(img. features = get_canny_features(img. print classification_report(features, labels=['spotted','not spotted']. preds = self.predict(features. predictedLabelSet = [] for pred in preds. predictedLabelSet += ['spotted'] + pred predictedLabelSet += ['not spotted'] + pred print print 'Classification Zone:' print classification_report(predictedLabelSet, labels=['spotted','not spotted']. predictedLabelSet = [] for pred in preds. predictedLabelSet += ['spotted'] + pred predictedLabelSet += ['not spotted'] + pred blur = cv2.GaussianBlur(preds, (20, 20), 0.astype('uint8'. print classification_report(predictedLabelSet, labels=['spotted','not spotted']. rawPreds = [] for pred in predictedLabelSet. rawPreds += [pred] features = get_canny_features(blur. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. probabilites = get_probabilities(features, rawPreds. print predictionScoresScatterplot(. print precisionRecallCurve(. print confusionMatrix(. def predictionScoresScatterplot(). predResultScatterplotResults=[ ] for i in range (0 , len (predictedLabelSet )). xscore , yscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore , zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ,zscore ] print 'Score distribution:' print 'The relative score for each class:' for i in range (0 , len (predictedLabelSet )). poslabel , neglabel = labels [ i ] poslabel += "+" neglabel += "-".join("+" if p == 1 else "-" if p == 0 else " " . xypospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospospos posnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegnegneg neg xy xy xy xy xy xy xy xy xy xy xy xy xy xy xy xy x

The process to integrate Canny and ShipRocket may seem complicated and intimidating. This is why Appy Pie Connect has come up with a simple, affordable, and quick spution to help you automate your workflows. Click on the button below to begin.