Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 14 Next »

Target release

January 2024

Document status

DRAFT

Document owner

@ mention owner

Designer

@ designer

Tech lead

@ lead

The Image Aquisition System will be a payload on the BenchBot and this document will not re-state requiermernts for the BenchBot Bench Bot v3 Concept Design - BenchBot - Confluence (atlassian.net)

Success metrics

Goal: Automated high throughput collection of high resolution images and associated metadata.

Objectives:

  • Collect high resolution images in an automated and systematic manner.

  • Collect environmental metadata.

  • Track metadata corresponding to each image.

  • Automate data upload.

  • Obtain camera position/orientation relative to grid.

  • Good user experience.

Table of Contents

Introduction

Image Quality

IQ.1.0 - Image quality

IQ.2.0 - Plant size

IQ.3.0 - Minimize adjustments

IQ.4.0 - Visual feedback

Sensor

S.1.0 - Cage

S.2.0 - Power

S.3.0 - Mounting to Gantry

S.2.0 - Image throughput (edge)

S.2.0 - Image throughput (cloud)

Image Capture

IC.1.0 - Y slider

Speed and Throughput

ST.1.0 - Image Capture

ST.2.0 - Motion

Metadata

M.1.0 - Species ID

M.2.0 - Weather info

M.3.0 - Weather station

M.4.0 - Incident light sensor

M.5.0 - RTK

M.6.0 - Position information

M.7.0 - IMU data

Requirement Workflow

Introduction

The purpose of BB 3.0 is to collect large amounts of high throughput image data in semi-controlled outdoor nursery conditions of agriculturally relevant weeds, cover crops, and cash crops. Images are collected in such a way that they can be semi-automatically annotated, enabling the development of a comprehensive image repository. The image repository will serve as a valuable resource for computer vision researchers and professionals in the field.

Image Quality Requirements

Id

Requirement name

Description

Category

Allowable Limits

Status

1.1

Image quality

Shall exceed state-of-the-art image quality standards to align with future use cases and camera systems.

Image Quality

1.2

Image Resolution

Shall meet the allowable limits for GSD (Ground Sample Distance)

Image Quality

9.1 pixels per mm (emergence)

5.9 pixels per mm (late growth stage)

Minimum resolution

1.3

MP

Image Quality

1.4

Sensor size

Full frame

Image Quality

1.6

Sensor type

CMOS

Image Quality

1.7

Pixel size

Image Quality

1.8

File type

Shall output RAW and JPG files

Image Quality

IQ 3.1

File architecture

Shall create a folder per data collection day and save files in such folder

IQ.2.0

Plant size

Shall support plant sizes with a height ranging from 0 cm to 40 cm above the pot soil

IQ.3.0

Automation

Shall minimize day to day adjustments. These include aperture control during image collection, and preprocessing and white balancing in post

IQ.4.0

Visual feedback

Should provide visual feedback to the user for manual inspection of errors during data collection

User Interface

Auto focus

Image throughput

Shall be able to capture images of 500 pots in less than 4 hours (500 images)

Speed and Throughput

Data transfer speed

Shall be such that does not jeopardize the minimum required image throughput.

Speed and Throughput

Species ID

Shall provide species ID for each pot in X, Y position

Metadata

Weather info

Shall have weather data information for each run captured from weather station

Metadata

Weather station

Shall have a weather station

Metadata

Light sensor

Shall have incident light sensor positioned near camera

Metadata

Shall allow the user to enter metadata

User Interface

Shall allow the user to share data

User Interface

Shall allow the user to control the bot

User Interface

Shall allow the user to change camera settings

User Interface

Shall allow the user to trigger camera independently

User Interface

Camera position

Shall provide camera position and orientation

Metadata

+/- 1 cm

+/- 1o

Emergent design properties

Camera position

RTK

Shall have individual images tagged with RTK position information (x,y,z)

Camera position

IMU

Shall have IMU data for camera position at each capture event

Sensor Requirements

Key

Summary

Specification

Allowable Limits

Status

S.1.0

Auto focus

should be designed to

S.2.0

S.3.0

S.4.0

Image Capture Requirements

Key

Summary

Specification

Allowable Limits

Status

IC.1.0.0

Y position

Shall have the ability of the Y camera slider to stop in certain positions

Speed and Throughput Requirements

Key

Summary

Specification

Allowable Limits

Status

ST.1.0.0

Image throughput

ST.2.0.0

Data transfer speed

Metadata Requirements

Key

Summary

Specification

Allowable Limits

Status

M.1.0

Species ID

Shall provide species ID for each pot in X, Y position

M.2.0

Weather info

Shall have weather data information for each run captured from weather station

M.3.0

Weather station

Shall have a weather station

M.4.0

Light sensor

Shall have incident light sensor positioned near camera

M.5.0

RTK

Shall have individual images tagged with RTK position information (x,y,z)

+/- 3 cm

M.6.0

IMU

Shall have IMU data for camera position at each capture event

M.7.0

Assumptions

  • This will be mounted to the end effector of an X, Y, Z gantry on the new BenchBot.

  • The BenchBot will be able to position this system to an accuracy of +/- 1 cm.

  • This system will request positions from the BenchBot and can trust it was accurately reached.

(warning) Out of Scope

  • No labels

0 Comments

You are not logged in. Any changes you make will be marked as anonymous. You may want to Log In if you already have an account.