SOPHON ACADEMY / Task
Introduction

Introduction

This task is designed to develop a QR code detection and recognition algorithm on the CV180X/CV181x processors for automatically identifying QR codes in images. The algorithm requires efficiently detecting the position of QR codes in images and accurately recognizing their content. It can be applied in various fields such as logistics, payment, advertising, etc., to achieve automated QR code scanning and information extraction.

 

  • Acceptance Criteria

The performance of the algorithm will be evaluated on a test set, requiring an accuracy of no less than 95% and FLOPS not exceeding 25G.

 

  • Test Set Conditions Description

The test set will include multiple scenarios to simulate various situations that may be encountered in actual applications. The specific requirements are as follows:

  1. Distance range: QR code images at different distances, for example, from 0.1 meters to 0.5 meters.
  2. Angle variation: Images at different angles, with the angle between the camera's optical center within -45 degrees to +45 degrees.
  3. Lighting conditions: Images under uniform lighting and low light conditions.
  4. Sample quantity: The test set should contain more than 500 samples to ensure a thorough assessment of the algorithm's performance.

With these requirements for the test set, we expect the algorithm to demonstrate high accuracy and robustness in actual application scenarios, meeting the real needs of QR code detection and recognition.

 

  • Submission Requirements
  1. Algorithm code adapted for the CV180X/CV181X processors.
  2. The test set (including data collection instructions) and an executable evaluation script for performance evaluation on the test set.
  3. Algorithm documentation, including detailed data on the algorithm's principles, model architecture, and performance (including accuracy, inference time, and computational complexity).

 

Data Collection Process

 

  • Data Collection Plan

The collection plan should be based on the actual application scenarios of the algorithm. First, determine the collection variables, such as the collection scene, quantity, number of people, gender, age, etc. Index and assign an English abbreviation to each variable. For example, if there are three variables: collection scene, distance, and gender, they could be numbered as:

  1. Scene 1: indoor; Scene 2: outdoor
  2. Distance 1: 1m; Distance 2: 3m; Distance 3: 5m
  3. Gender 1: male; Gender 2: female

Then, based on the variables, determine the collection process, such as whether to collect according to variable 1 or variable 2 first, the actions of the people being collected, and precautions, etc. Organize the above collection plan and variable information into a Word document for saving.

 

  • Prepare Collection Equipment

The collection equipment should be as close as possible to the actual equipment used and ensure that it can be preserved for a long time. Prepare the collection firmware.

 

  • Data Collection and Saving

Carry out data collection and name the files according to the order of the collection variables defined in the "Data Collection Plan," with names like x-x-x-x.xxx. For example, if there are three variables: scene, distance, and gender, the saving format would be: indoor-3m-female-12.xxx, representing the 12th data of "indoor scene, 3m distance, female" collected. After the collection is complete, the collected data and the collection instruction document from the "Data Collection Plan" should be put together for inspection.