Skip to content

Latest commit

 

History

History
38 lines (24 loc) · 1.31 KB

README.md

File metadata and controls

38 lines (24 loc) · 1.31 KB
Raw

CookingAlly

Easy recipes finder using real-time object detection on Android.

Overview

This demo app allows users to find quick recipes using their camera. The object detection model utilizes YOLOv5 and Tensorflow Lite.

Development Environment

  • Roboflow
  • Yolov5
  • Tensorflow 1.15.2
  • Android Studio Giraffe

The custom dataset

Number of classes: 5 (carrot, cheese, egg, potato, zucchini)
Number of images: 2360
Split: Train set 75%, Valid set 16%, Test set 9%

Our custom dataset can be found at https://universe.roboflow.com/project-0tja1/ingredients-cfeum

The app

There are four main activities:

  • CameraActivity: manages interactions with the device's camera.
  • Detector Activity: uses the TensorFlow API to detect objects in an image.
  • MainActivity: incorporates the core logic of the application, including reading recipes from an Excel file, selecting recipes based on detected ingredients and creating the log.
  • ViewRecipeActivity: handles the organized display of recipe features, including the associated image.

Usage

To build the pretrained demo in Android Studio, import the project and navigate to the cookingAlly/cookingAlly directory. Then, connect a device, frame the ingredients, and view the recommended recipes.