1. Data collection and preparation: Collect images and related annotation data including faces. The annotation data can include information such as face frame position and face feature points. This data will be used to train and evaluate the model. 2. Data preprocessing: preprocessing the collected images, including image scaling, cropping, brightness and contrast adjustment, etc. These steps are designed to improve the performance and robustness of the model. 3. Model training: Using the YOLO algorithm as the basis, using the prepared data set to train the face recognition model. The goal of training is to enable the model to accurately detect and recognize faces. 4. Model evaluation and tuning: evaluation data sets are used to evaluate the trained model and calculate the performance indicators of the model in face detection and recognition tasks, such as accuracy rate and recall rate. According to the evaluation results, the model is optimized and improved. 5. Deployment and application: Deploy the trained model to the actual application environment, such as video surveillance system, face recognition access control system, etc. Ensure that the model can detect and recognize faces in real time and accurately.

Realizing recognition of 120 different categories dog with tflearn and as possible as we can optimize the model. Completing the user-oriented graphic interface,with giving the feedback and analysis results based on the dog pictures given by users.

Designed and implemented takeaway catering order system, applied the Maven to develop program functions including food inquiry, good management, food editing, errand boy management, and ordering records, etc. Used technology of SpringBoot, Vue, and Python, applied development tools of IDEA, WebStorm, PyCharm, DataGrip，finished the code design, used the SpringMVC mode lightweight framework for development and Python to acquire data