Water Object Identification Using Neural Image Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for detecting objects in water, such as sonar, radar, and lidar technologies, are inadequate for identifying objects partially or fully submerged, especially at distances beyond 200 meters, and fail to differentiate between stationary and moving objects, posing a risk for boat collisions.
Innovation Solution
A system utilizing a camera module and an artificial neural network to generate and process images of water areas, detecting and identifying objects by matching extracted features with predefined classes, and providing navigation data to avoid collisions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sonar-based technologies are used to detect immersed objects, then detection capability for submerged objects is improved, but the system cannot detect floating objects that are not significantly immerged
Solution Approach 1:
The patent combines multiple detection technologies (sonar for submerged objects, radar/lidar for floating objects) into a single integrated system. The processor coordinates these different detection modules to provide comprehensive coverage for both immersed and floating objects, resolving the contradiction between specialized detection capability and overall adaptability.
2Length of stationary object
If radar systems are used to detect floating objects, then detection range is improved, but the system cannot identify object types or differentiate between moving and stationary objects
Solution Approach 1:
The patent introduces an artificial neural network as an intermediary processing layer that analyzes data from radar and lidar systems. This neural network extracts additional information about object types, movement patterns, and characteristics, converting raw detection data into identifiable object information while maintaining long-range detection capability.
3Measurement precision
If lidar systems are used to detect floating objects, then object detection accuracy is improved, but identification is limited to distances below 200 meters due to eye safe laser power limitations
Solution Approach 1:
The patent creates a multi-functional detection system where radar provides long-range initial detection, lidar provides detailed close-range analysis, and an artificial neural network integrates both data sources. This universal system achieves both long-range capability (through radar) and high-accuracy identification (through lidar and neural network processing) at various distances.
4Speed
If signal-based detection systems are used, then detection speed is improved, but signals may be deviated or modified during transmission causing detection failures
Solution Approach 1:
The patent uses optical copying (imaging) instead of signal-based detection. Cameras capture visual images of objects, creating a direct optical copy of the target. This approach maintains fast detection speed while improving reliability, as optical images are not subject to the same deviation and modification issues as acoustic or electromagnetic signals in complex environments.
Data Source
AI summary
The invention relates to a system for identifying at least one object at least partially immerged in a water area, said system comprising a capturing module comprising at least one camera, said at least one camera being configured to generate at least one sequence of images of said water area, and a processing module being configured to receive at least one sequence of images from said at least one camera and comprising at least one artificial neural network, said at least one artificial neural network being configured to detect at least one object in said at least one received sequence of images, extract a set of features from said at least one detected object, compare said extracted set of features with at least one predetermined set of features associated with a predefined object, identify the at least one detected object when the extracted set of features matches with the at least one predetermined set of features.