{"id":20672216,"url":"https://github.com/iswarya-singaram/object_detection_using_tensorflowlite","last_synced_at":"2026-04-30T16:31:27.944Z","repository":{"id":253215960,"uuid":"813581781","full_name":"Iswarya-Singaram/Object_Detection_Using_TensorflowLite","owner":"Iswarya-Singaram","description":"This project leverages the power of machine learning to identify objects in real-time through a web camera. Designed to be lightweight and efficient, it's a perfect demonstration of edge AI capabilities.","archived":false,"fork":false,"pushed_at":"2024-09-28T14:08:13.000Z","size":22953,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-02-21T02:24:15.091Z","etag":null,"topics":["github","raspberry-pi","tensorflow-lite"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Iswarya-Singaram.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null}},"created_at":"2024-06-11T10:52:28.000Z","updated_at":"2024-10-02T03:37:18.000Z","dependencies_parsed_at":"2025-03-10T17:46:49.280Z","dependency_job_id":null,"html_url":"https://github.com/Iswarya-Singaram/Object_Detection_Using_TensorflowLite","commit_stats":null,"previous_names":["iswarya-singaram/object_detection_using_tensorflowlite"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Iswarya-Singaram/Object_Detection_Using_TensorflowLite","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Iswarya-Singaram%2FObject_Detection_Using_TensorflowLite","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Iswarya-Singaram%2FObject_Detection_Using_TensorflowLite/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Iswarya-Singaram%2FObject_Detection_Using_TensorflowLite/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Iswarya-Singaram%2FObject_Detection_Using_TensorflowLite/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Iswarya-Singaram","download_url":"https://codeload.github.com/Iswarya-Singaram/Object_Detection_Using_TensorflowLite/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Iswarya-Singaram%2FObject_Detection_Using_TensorflowLite/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":32470879,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-04-30T13:12:12.517Z","status":"ssl_error","status_checked_at":"2026-04-30T13:12:06.837Z","response_time":57,"last_error":"SSL_read: unexpected eof while reading","robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":false,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["github","raspberry-pi","tensorflow-lite"],"created_at":"2024-11-16T20:32:52.543Z","updated_at":"2026-04-30T16:31:27.930Z","avatar_url":"https://github.com/Iswarya-Singaram.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Object Detection Using Tensorflow-Lite\nWelcome to our Object Detection project, built using TensorFlow Lite and Raspberry Pi! This project leverages the power of machine learning to identify objects in real-time through a web camera. Designed to be lightweight and efficient, it's a perfect demonstration of edge AI capabilities, making it ideal for various applications such as home automation, security, and educational purposes.This guide provides step-by-step instructions for how to set up TensorFlow Lite on the Raspberry Pi and use it to run object detection models\n\u003cp align=center\u003e\n\u003cimg src=\"https://github.com/user-attachments/assets/62a2207d-438f-4402-9cb8-b64190d89a65\" width=\"500\" height=\"300\"\u003e\n\u003c/p\u003e\n\n## Features\n\u003cul\u003e\n\u003cli\u003eReal-Time Object Detection: Identify and classify objects in real-time using a live feed from a web camera.\u003c/li\u003e\u003c/br\u003e\n\u003cli\u003eEdge Computing: Efficiently runs on Raspberry Pi, showcasing the potential of edge AI devices.\u003c/li\u003e\u003c/br\u003e\n\u003cli\u003eTensorFlow Lite: Utilizes TensorFlow Lite for optimized performance on low-power devices.\u003c/li\u003e\u003c/br\u003e\n\u003cli\u003eEasy Setup: Simple installation and configuration process to get the model up and running quickly.\u003c/li\u003e\u003c/br\u003e\n\u003c/ul\u003e\n\u003cp align=center\u003e\n\u003cimg src=\"https://github.com/user-attachments/assets/20478218-31d0-462a-95b3-d9ce8eeef9c9\" width=\"500\" height=\"300\"\u003e\n\u003c/p\u003e\n\n## How it works\n\u003cul\u003e\n\u003cli\u003eCapture: The Raspberry Pi's connected web camera captures images in real-time.\u003c/li\u003e\n\u003cli\u003eProcess: The captured images are processed using a pre-trained TensorFlow Lite object detection model.\u003c/li\u003e\n\u003cli\u003eIdentify: The model identifies the objects present in the images and outputs their names.\u003c/li\u003e\n\u003c/ul\u003e\n\n\n ## Requirements\n   \u003col\u003e\n\u003cli\u003eRaspberry Pi 4 or 5 (preferably with 2GB RAM or more)\u003c/li\u003e\n\u003cli\u003eWeb camera compatible with Raspberry Pi\u003c/li\u003e\n\u003cli\u003eTensorFlow Lite\u003c/li\u003e\n\u003cli\u003ePython 3\u003c/li\u003e\n \u003c/ol\u003e\n\n ## Getting Started\n The Getting Started part consists of the following subparts \n \u003col\u003e\n   \u003cli\u003eSetting up the raspberry pi\u003c/li\u003e\n   \u003cli\u003eDownload the github repository and create an environment\u003c/li\u003e\n   \u003cli\u003eInstalling all the necessary packages\u003c/li\u003e\n   \u003cli\u003eSetting up the object detection model\u003c/li\u003e\n   \u003cli\u003eRunning the model\u003c/li\u003e\n \u003c/ol\u003e\n\n ## Setting Up the Raspberry Pi\n ### Step-1:Update Your Raspberry Pi\n Open the teminal in your raspberry pi and update it using the following command\n ~~~\nsudo apt-get update\nsudo apt-get dist-upgrade\n~~~\n The upgrade process should only take a minute or two, depending on when it was last updated.\n ### Step-2:Set Up the Configuration\nFor Raspberry Pi 4 and 5:\nNo additional configuration is needed, as the camera is enabled by default.\n\nFor Lower Versions of Raspberry Pi:\n\u003col\u003e\n\u003cli\u003eClick the Pi icon in the top left corner of the screen.\u003c/li\u003e\n\u003cli\u003eSelect Preferences -\u003e Raspberry Pi Configuration.\u003c/li\u003e\n\u003cli\u003eGo to the Interfaces tab.\u003c/li\u003e\n\u003cli\u003eVerify that Camera is set to Enabled. If it is not, enable it now.\u003c/li\u003e\n\u003cli\u003eReboot the Raspberry Pi to apply the changes.\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/Iswarya-Singaram/Object_Detection_Using_TensorflowLite/assets/145309713/7dc9ce06-914d-4be7-9e5f-30a3b0448817\"\u003e\n\u003c/p\u003e\n\n## Download the github repository and create an environment\n### Step1: Download the github repository\nNext, clone the github repository by using the following command\n~~~\ngit clone https://github.com/EdjeElectronics/TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi.git\n~~~\nThis downloads everything into a folder called TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi.Since, the folder name is too long, we can rename it to tflite1 by using the following command. Then use cd to change the directory to tflite1\n~~~\nmv TensorFlow-Lite-Object-Detection-on-Android-and-Raspberry-Pi tflite1\ncd tflite1\n~~~\n### Step 2:Create an environment\n Next up is to create a virtual environment called \"tflite1-env\".\n Install virtualenv by using\n ~~~\nsudo pip3 install virtualenv\n~~~\nThen, create the \"tflite1-env\" virtual environment by issuing:\n~~~\npython3 -m venv tflite1-env\n~~~\nThis will create a folder called tflite1-env inside the tflite1 directory. The tflite1-env folder will hold all the package libraries for this environment. Next, activate the environment by issuing:\n~~~\nsource tflite1-env/bin/activate\n~~~\nNote:To reactivate the tflite1-env virtual environment every time you open a new terminal window, you should navigate to the /home/pi/tflite1 directory and issue the command source tflite1-env/bin/activate. When the virtual environment is active, you will see (tflite1-env) appear at the beginning of your command prompt.\n## Installing all the necessary packages\nNext, we'll install TensorFlow and OpenCV along with their required dependencies. While OpenCV is not essential for running TensorFlow Lite, the object detection scripts in this repository utilize it to capture images and visualize detection results.\nTo install opencv use,\n~~~\npip install opencv-python\n~~~\nNow, install tensorflow\n~~~\npip install tensorflow\n~~~\nYou can install any tensorflow version of your choice\n\u003cp align=\"center\"\u003e\n  \u003cimg src=\"https://github.com/Iswarya-Singaram/Object_Detection_Using_TensorflowLite/assets/145309713/6fc1c8f5-24c0-41c0-a6c6-3d9cc734e976\" width=\"500\" height=\"300\"\u003e\n\u003c/p\u003e\nIf you come across an error like \"Failed to build wheel for h5py\" as shown in the screenshot above, follow these steps in the terminal...\nInstall h5py by cloning the git\n\n~~~\ngit clone https://github.com/h5py/h5py.git\n~~~\nAnd change the directory into h5py\n~~~\ncd h5py\n~~~\nNow install libhdf5,protobuf,Cython,pkgconfig using the following commands\n~~~\nsudo apt install python3-dev libhdf5-dev\npip install protobuf==3.20\npip install Cython\npip install pkgconfig\n~~~\nNow install setup.py using the following command\n~~~\npython setup.py install\n~~~\nWe can leave the  h5py directory but using the command\n~~~\ncd ..\n~~~\nNow, we can install tensorflow without having to worry about the error.\n~~~\npip install tensorflow\n~~~\n## Setting up the object detection model\nDownload the sample model (which can be found on the Object Detection page of the official TensorFlow website) by issuing:\n~~~\nwget https://storage.googleapis.com/download.tensorflow.org/models/tflite/coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip\n~~~\nUnzip it to a folder called \"Sample_TFLite_model\" by issuing (this command automatically creates the folder):\n~~~\nunzip coco_ssd_mobilenet_v1_1.0_quant_2018_06_29.zip -d Sample_TFLite_model\n~~~\n## Running the model\nNow run the Object detection model using the following command\n\n~~~\npython3 TFLite_detection_webcam.py --modeldir=Sample_TFLite_model\n~~~\n\n## Credits\nThis project is based on the work done by Edje Electronics. We extend our gratitude for their comprehensive guide on TensorFlow Lite Object Detection.\n\n## Contributing\nWe welcome contributions to improve this project. If you find any errors in the code or have suggestions for improvements, please feel free to open an issue or submit a pull request. Your feedback and contributions are greatly appreciated!\n\n\n\n\n\n\n\n\n \n\n\n\n \n\n \n \n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiswarya-singaram%2Fobject_detection_using_tensorflowlite","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fiswarya-singaram%2Fobject_detection_using_tensorflowlite","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fiswarya-singaram%2Fobject_detection_using_tensorflowlite/lists"}