LSM-RNN Object Recognition for Noisy Image Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The performance of machine learning algorithms can be adversely affected by noise or broken image data, leading to variations in recognition rates, necessitating improved data preprocessing methods to minimize processing and enhance performance.
Innovation Solution
An electronic apparatus utilizing a liquid-state machine (LSM) model and recurrent neural networks (RNN) model for preprocessing input data, where the LSM model processes feature data and adjusts weights based on neuron spike activity to optimize the input for the RNN model, which is trained on preset objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a camera module is installed in a smartphone to enable photographing functions, then the smartphone gains photographing capability, but the smartphone's thickness increases
Solution Approach 1:
The flash assembly is nested within the cavity of the camera module, and the camera module is integrated into the smartphone body such that the combined height of these components does not exceed the thickness of the smartphone. This nesting arrangement allows the smartphone to gain both photographing capability and flash functionality without increasing its overall thickness.
2Object-affected harmful factors
If the distance between the light source and the illuminating object is increased to reduce the flash's impact on the user, then the flash effect is reduced, but the illuminating area expands
Solution Approach 1:
The reflecting plate is positioned adjacent to the lens, creating a localized reflection path that directs light precisely onto the illuminating object. This local quality approach ensures that while the light source remains at a distance from the object (reducing flash impact on user), the reflected light maintains sufficient intensity and focus to achieve reliable illumination of the target area.
3Object-affected harmful factors
If the reflecting plate is positioned far from the lens, then the flash's direct impact on the user is reduced, but the illuminating area becomes insufficient
Solution Approach 1:
The reflecting plate acts as an intermediary element that receives light from the light source and redirects it toward the illuminating object. By positioning the reflecting plate adjacent to the lens rather than far from it, the system maintains an effective light path that provides sufficient illuminating area while still keeping the light source at a distance from the user to reduce flash impact.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The LSM-RNN combination effectively identifies preset objects in input data by enhancing the recognition rate and minimizing data processing, improving the performance of machine learning algorithms in noisy or degraded input conditions.
Implementation Method 1
a reflecting plate adjacent to the lens in the camera module, the light source being disposed in the flash assembly at a position opposite to the reflecting plate
Data Source
Figure 1A~1B
Figure 2
Figure 3
AI summary
An electronic apparatus includes a storage configured to store a liquid-state machine (LSM) model and a recurrent neural networks (RNN) model, and a processor configured to input and process a feature data acquired from an input data using the LSM model, to input and process an output value output by the LSM model using the RNN model, and to identify whether a preset object is included in the input data based on an output value output by the RNN model. The RNN model is trained by a sample data related to the preset object. The LSM model includes a plurality of interlinked neurons. A weight applied to a link between the plurality of interlinked neurons is identified based on a spike at which a neuron value is greater than or equal to a preset threshold in a preset unit time.