情绪识别-基于OpenCV+Deepface实现的人脸面部情绪识别算法-优质项目实战.zip

# Facial-Emotion-Recognition-using-OpenCV-and-Deepface ## Dependencies - [deepface](https://github.com/serengil/deepface): A deep learning facial analysis library that provides pre-trained models for facial emotion detection. It relies on TensorFlow for the underlying deep learning operations. - [OpenCV](https://opencv.org/): An open-source computer vision library used for image and video processing. ## Usage ### Initial steps: - Download this repository - Run: `cd 项目` 1. Install the required dependencies: - You can use `pip install -r requirements.txt` - Or you can install dependencies individually: - `pip install deepface` - `pip install opencv-python` 2. Download the Haar cascade XML file for face detection: - Visit the [OpenCV GitHub repository](https://github.com/opencv/opencv/tree/master/data/haarcascades) and download the `haarcascade_frontalface_default.xml` file. 3. Run the code: - Execute the Python script. - The webcam will open, and real-time facial emotion detection will start. - Emotion labels will be displayed on the frames around detected faces. ## Approach 1. Import the necessary libraries: `cv2` for video capture and image processing, and `deepface` for the emotion detection model. 2. Load the Haar cascade classifier XML file for face detection using `cv2.CascadeClassifier()`. 3. Start capturing video from the default webcam using `cv2.VideoCapture()`. 4. Enter a continuous loop to process each frame of the captured video. 5. Convert each frame to grayscale using `cv2.cvtColor()`. 6. Detect faces in the grayscale frame using `face_cascade.detectMultiScale()`. 7. For each detected face, extract the face ROI (Region of Interest). 8. Preprocess the face image for emotion detection using the `deepface` library's built-in preprocessing function. 9. Make predictions for the emotions using the pre-trained emotion detection model provided by the `deepface` library. 10. Retrieve the index of the predicted emotion and map it to the corresponding emotion label. 11. Draw a rectangle around the detected face and label it with the predicted emotion using `cv2.rectangle()` and `cv2.putText()`. 12. Display the resulting frame with the labeled emotion using `cv2.imshow()`. 13. If the 'q' key is pressed, exit the loop. 14. Release the video capture and close all windows using `cap.release()` and `cv2.destroyAllWindows()`.