Agricultural Vehicle Camera Image Retrieval by Time and Position

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing systems for electronically managing agricultural crop cultivation struggle with capturing images of specific targets, such as equipment failures or weeds, while a worker is on a moving vehicle, as the timing of target detection does not align with the camera's field of view, leading to unnecessary image capture and increased effort in identifying relevant images.

Innovation Solution

An information processing apparatus that chronologically manages images from a camera mounted on a moving object, allowing users to input instructions for specific report content, and identifies suitable images based on time and position information, using a candidate image group selection and object detection techniques to quickly retrieve relevant images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of operation

If a camera is mounted on a moving vehicle to capture images while the worker is on the vehicle, then the worker does not need to get off the vehicle to image incidents, but the timing of target detection does not align with the camera's field of view, leading to missed targets

Engineering Contradiction:
Improveworker convenienceVSAvoidtarget capture reliability
Core Design Contradiction:
Ease of operationVSReliability

Solution Approach 1:

The system performs preliminary actions by continuously capturing images and storing them with timestamp and position information before the worker issues an instruction. When the worker discovers a target and inputs an instruction, the system can retrieve images captured around that time, ensuring the target is captured even though the worker did not trigger the camera at the exact moment of discovery.

Inventive Principle:
Principle #10Preliminary action

2Reliability

If the camera constantly captures images to prevent missing targets, then all targets can be captured, but the number of recorded images increases considerably, requiring more time and effort to identify relevant images

Engineering Contradiction:
Improvetarget capture completenessVSAvoidimage identification time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system extracts only the necessary information from the large set of captured images by using timestamp and position data. When a worker issues an instruction, the system extracts and displays only the images captured around the instruction time, filtering out the vast majority of irrelevant images and enabling quick identification of the target.

Inventive Principle:
Principle #2Taking out (Extraction)

3Quantity of substance

If images are captured only when instructed by the worker, then the number of recorded images is reduced, but the target may not be within the camera's angle of view at the instruction timing

Engineering Contradiction:
Improvenumber of recorded imagesVSAvoidtarget positioning accuracy
Core Design Contradiction:
Quantity of substanceVSMeasurement precision

Solution Approach 1:

The system performs preliminary image capture and stores images with timestamp and position information before the worker issues an instruction. This allows the system to retrieve images captured just before or after the instruction time, ensuring the target is captured even if it moves in or out of the camera's field of view between the worker's discovery and the instruction.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS11386651B2Information processing apparatus that manages image captured at site where agricultural crop is cultivated, method for controlling the same, storage medium, and system
Publication Date: 2022.07.12 CANON KK
  • US11386651B2 patent drawing
  • US11386651B2 patent drawing
  • US11386651B2 patent drawing

AI summary

An information processing apparatus according to an aspect of the present invention displays a plurality of items corresponding to a plurality of kinds of predefined targets on a display device, manages a plurality of images, each in association with information regarding a position and a time at which each of the plurality of images is captured, chronologically captured by an imaging apparatus mounted on a moving object, and acquires an image containing a target corresponding to a selected item among the plurality images based on information regarding a position of the moving object or the imaging apparatus when an instruction based on a selection of any of the plurality of displayed items is input, information regarding a time at which the instruction is input, and the selected item.