OpenCV 文字檢測與識別模塊

OpenCV 文字檢測與識別模塊

該模塊在擴展模塊中，需自行下載
下載地址：https://github.com/opencv/opencv_contrib/tree/4.0.0
說明文檔：
文字檢測 https://docs.opencv.org/4.0.0/da/d56/group__text__detect.html
文字識別 https://docs.opencv.org/4.0.0/d8/df2/group__text__recognize.html
參考文章：https://github.com/opencv/opencv_contrib/blob/master/modules/text/samples/text_recognition_cnn.cpp
https://github.com/opencv/opencv_contrib/blob/master/modules/text/samples/webcam_demo.cpp

文字檢測

1.textDetectorCNN

OpenCV的文字檢測模塊textDetectorCNN中使用了TextBoxes：具有單個深度神經網絡的快速文本檢測器 鏈接地址爲： https://github.com/MhLiao/TextBoxes
其中已經訓練過的文件：

函數名	內容	地址
modelWeightsFilename	描述分類器體系結構的prototxt文件的相對或絕對路徑。	textbox.prototxt 在下載的擴展模塊源碼中opencv_contrib/modules/text/samples/textbox.prototxt
modelWeightsFilename	包含caffe-binary形式的模型的預訓練權重的文件的相對或絕對路徑。	TextBoxes_icdar13.caffemodel http://pan.baidu.com/s/1qY73XHq

        cv::Mat temp;
		src.convertTo(temp, CV_8UC3, 1);//src 輸入圖像
		cv::imshow("src", temp);
		dst1 = temp.clone();
		cv::Ptr<cv::text::TextDetectorCNN> detector= cv::text::TextDetectorCNN::create("textbox.prototxt", "TextBoxes_icdar13.caffemodel");
		std::vector < cv::Rect > boxes;//識別區域
		std::vector < float > sources;//評估分數
		detector->detect(temp, boxes, sources);
		float threshold = 0.5;
		for (int i = 0; i < boxes.size(); i++)
		{
			if (sources[i] > threshold)
			{
				cv::Rect rect = boxes[i];
				cv::rectangle(dst1, rect, cv::Scalar(255, 0, 0), 2);
			}
		}
		cv::imshow("Text detection result", dst1);

原圖

結果

2.erGrouping

文字識別

1.OCRHolisticWordRecognizer

OCRHolisticWordRecognizer類提供了分段詞語的功能。給定預定義的詞彙表，使用DictNet來選擇給定輸入圖像的最可能的詞。

DictNet詳細描述於：Max Jaderberg等：使用卷積神經網絡閱讀野外文本，IJCV 2015 http://arxiv.org/abs/1412.1842
模型文件下載地址： http://nicolaou.homouniversalis.org/assets/vgg_text/dictnet_vgg.caffemodel
http://nicolaou.homouniversalis.org/assets/vgg_text/dictnet_vgg_deploy.prototxt
http://nicolaou.homouniversalis.org/assets/vgg_text/dictnet_vgg_labels.txt

wordSpotter = (cv::text::OCRHolisticWordRecognizer::create("dictnet_vgg_deploy.prototxt", "dictnet_vgg.caffemodel", "dictnet_vgg_labels.txt"));
dst1 = src.clone();
		for (size_t i = 0; i < textBoxes.size(); i++)
		{
			cv::Mat wordImg;
			cv::cvtColor(src(textBoxes[i]), wordImg, cv::COLOR_BGR2GRAY);
			std::string word;
			std::vector<float> confs;
			wordSpotter->run(wordImg, word, NULL, NULL, &confs);//檢測
			cv::Rect currrentBox = textBoxes[i];
			cv::rectangle(dst1, currrentBox, cv::Scalar(0, 255, 255), 2, cv::LINE_AA);
			int baseLine = 0;
			cv::Size labelSize = cv::getTextSize(word, cv::FONT_HERSHEY_PLAIN, 1, 1, &baseLine);
			int yLeftBottom = currrentBox.y>labelSize.height? currrentBox.y: labelSize.height;
			cv::rectangle(dst1, cv::Point(currrentBox.x, yLeftBottom - labelSize.height),
				cv::Point(currrentBox.x + labelSize.width, yLeftBottom + baseLine), cv::Scalar(255, 255, 255), cv::FILLED);
			cv::putText(dst1, word, cv::Point(currrentBox.x, yLeftBottom), cv::FONT_HERSHEY_PLAIN, 1, cv::Scalar(0, 0, 0), 1, cv::LINE_AA);
		}
		cv::imshow("Text recognition", dst1);		

結果