opencv+CNN实现实时手写数字识别

Different from the common practice of MNIST image recognition using CNN algorithm, I apply Numpy and OpenCV to extract relevant features from each MNIST figure, and then trains Xgboost recognition model. After gradually adjusting parameters, the accuracy of the optimal model on the test set can reach 88%. In additon, since I've made extensive use of the broadcasting-mechanism of NumPy instead of loops when coding, the code can run at an excellent speed.

Real-Time Digit Recognition with TensorFlow and OpenCV This Python script uses TensorFlow and OpenCV for real-time digit recognition from a webcam feed. It trains a CNN model on the MNIST dataset and predicts digits in real-time from the video stream.

🤖PYTHON 手写体数字验证码识别器 | MNIST CNN 模型 Handwritten Digit CAPTCHA Recognizer 🔮 一个使用 TensorFlow/Keras 和 OpenCV 构建的 Python 项目，用于识别简单的手写数字验证码图片。 包含模型训练脚本 (ModelTraining.py) 和模型调用脚本 (Model_Identify.py)

By using opencv, Tensorflow module with python3, Handwritten numeral data will be automatically extracted from video by opencv extracting contour method. The MNIST CNN train model will predicted the number by the extracted data and the predicted number will be shown on the on the video.(as a text)

Learned & Implemented Object Detection & Tracking ,Image & Video Processing using OpenCV.Also implemented Deep Learning models(YOLO,CNN,MNIST Dataset) using Keras,TensorFlow

Advanced Computer Vision & Deep Learning: Complete implementations of blob detection (LoG/DoG/DoH), image processing pipelines, edge detection, histogram analysis, DCT compression, semantic segmentation, YOLO/Faster R-CNN object detection, and CNN classifiers for MNIST/CIFAR/Fashion-MNIST using PyTorch, TensorFlow, and OpenCV

PyTorch CNN trained on MNIST to classify handwritten digits (0–9). Includes a training notebook that saves digit_cnn.pth and an OpenCV app to draw digits on a canvas and get instant predictions.

A machine learning project that utilizes deep learning (CNN) to recognize handwritten digits from the MNIST dataset. Built with Python, TensorFlow/Keras, and OpenCV for preprocessing.

Deep Learning for Pattern Recognition: A master’s project that trains a CNN on MNIST digits, adapts it via transfer learning for Greek letter classification, and implements live video digit recognition using OpenCV. Built with PyTorch, Torchvision, and OpenCV.

✍️ Handwritten digit recognition using CNN trained on MNIST dataset. Upload an image or draw a digit - AI predicts it instantly with confidence score! Built with TensorFlow, OpenCV & Tkinter GUI.

This project is a Handwritten Digit Recognition System using a CNN model trained on the MNIST dataset. It includes a GUI built with Tkinter, where users can draw digits, and OpenCV processes the input for real-time classification.

This project use opencv and mediapipe to recognize hand. As hand snips, it can be drawn something on the screen. I trained the CNN which based on MNIST to recognize numbers, so this project can recognize the number written.

This project detects digits using CNNs and OpenCV. It includes image preprocessing, model training on MNIST, real-time detection, and evaluation with accuracy metrics. Built with Python, TensorFlow/Keras, and Flask for deployment. Clone, install dependencies, and start detecting digits! 🚀

A CNN-based handwritten digit recognition project using the MNIST dataset. Built with TensorFlow, Keras, OpenCV, and Jupyter Notebook. Users can draw digits (0–9) in a window, and the model predicts them in real-time. Simple, interactive, and beginner-friendly.

A real-time AI project that recognizes digits drawn in the air using your finger. Built with Python, TensorFlow, OpenCV, and MediaPipe — this system tracks hand motion, captures the drawn pattern, and predicts the digit using a CNN trained on the MNIST dataset.

Python code uses TensorFlow and OpenCV to implement a CNN for recognizing handwritten digits. It loads the MNIST dataset, preprocesses the images, defines and trains the model, and evaluates its performance, it predicts a digit from a sample handwriting image fetched from a URL.

CNN model that tries to predict digits in the written paper

No description available

keras mnist cnn with web(flask) and opencv-python demo

No description available

Real-time handwritten digit recognition using CNN (MNIST) + OpenCV live camera input.

Real-time handwritten digit recognition from webcam using MNIST CNN and OpenCV

CNN-based MNIST digit classifier with real-time recognition using OpenCV and webcam input.

An interactive drawing application built with OpenCV and PyTorch using a trained CNN based on MNIST.

Real-time handwritten digit recognition using a CNN trained on MNIST with webcam input (OpenCV + TensorFlow).

A Convolutional Neural Network (CNN) based Handwritten Digit Recognition using MNIST Dataset in Python with TensorFlow & OpenCV.

Handwritten Digit Recognition using CNN (PyTorch) with OpenCV drawing interface, achieving 98.89% accuracy on MNIST dataset (CSV format).

Drawing Digits through a customizable pointer using opencv and python and using a pretrained CNN model on MNIST to detect digits 0-9.

A CNN trained on the MNIST dataset used with OpenCV to recognize handwritten digits drawn in the air using a colored object tracked via webcam.