Introduction

Computer vision is a very interesting subject in the area of image analytics and a lot of use cases exist in this field. These are broadly classified as image classification, image object detection, image segmentation and image generation.

In this blog we will stick to image object detection. This topic has been gaining a lot of eye balls for commercial as well as research use as it deals around detecting objects in an image. For example detecting vehicles, animals, mobile phones, aircraft and any other object that you can think of. Object detection models are mostly used in self driving cars or fire detection mechanism or retail store analytics.

About the problem and Dataset:

We will be using Tensorflow’s Object Detection API on TF 2 to fine-tune existing Object Detection models for a custom dataset on Google Colab. There are a couple of blogs on internet for this but they are all on TF 1.X

**My special credits go to **Dat Tran **on all his effort for creating the Raccoon dataset and labeling them, **dataset is available on Github — https://github.com/datitran/raccoon_dataset.

The dataset contains 200 images of raccoons which were manually box-bound labeled using the tool LabelImg to prepare the dataset in the required format.

Object Detection approach:

The object detection workflow comprises of the below steps:

1. Collecting the images to train and validate the Object Detection model.
2. Labeling the dataset using a tool like LabelImg.
3. Preparing a TFRecord file for ingesting in object detection API.
4. Installing the Tensorflow Object Detection API.
5. Creating the OD Config file.
6. Running the Object detection training and eval job.
7. Exporting the model.

#tensorflow2 #object-detection-api #object-detection #tensorflow