Compressed Sensing Object Imaging System for Real-Time Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing target recognition methods suffer from poor real-time performance and large volumes of image data due to sequential processing of images, which limits their development and efficiency, especially in the context of information transmission and processing speed constraints.
Innovation Solution
A compressed sensing based object imaging system that performs image compression, sampling, and recognition in a pure optical domain, utilizing a light source generation unit, filter unit, image generation unit, and image reconstruction unit to reduce data volume and enable concurrent processing by machine vision and artificial intelligence, employing a specific measurement matrix trained using principal component analysis for image reconstruction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional target recognition methods process images sequentially after acquisition, then recognition can be performed, but the system real-time performance deteriorates and large volumes of image data are generated
Solution Approach 1:
The patent applies preliminary action by performing target recognition during the imaging process itself rather than after image acquisition. The measurement matrix is designed to encode target recognition information directly into the compressed sensing measurements, enabling recognition to occur concurrently with data acquisition. This eliminates the sequential processing bottleneck and improves real-time performance.
Solution Approach 2:
The patent merges image compression and target recognition into a single integrated process. By designing the measurement matrix to contain target-specific patterns, the system simultaneously compresses the image data and performs recognition, combining two previously separate operations into one concurrent process that improves both efficiency and real-time performance.
2Measurement precision
If the sampling rate is increased to obtain high-quality data, then data quality improves, but the volume of image data increases and storage and transmission become bottlenecks
Solution Approach 1:
The patent changes the parameter of sampling rate by using compressed sensing to acquire images at rates below the traditional Nyquist sampling rate. The measurement matrix is designed to capture essential information at lower sampling rates, maintaining data quality while significantly reducing the volume of acquired data and eliminating storage and transmission bottlenecks.
3Quantity of substance
If compressed sensing is used to reduce data volume, then data acquisition volume decreases, but cumbersome image information still exists during further image recognition and processing
Solution Approach 1:
The patent applies preliminary action by pre-designing the measurement matrix to encode target recognition information directly into the compressed measurements. This preliminary encoding of recognition patterns into the measurement process eliminates the need for subsequent complex image processing steps, reducing processing complexity while maintaining reduced data volume.
4Ease of manufacture
If digital signal processing is used for compressed sensing, then implementation is achieved, but the advantages of concurrent processing are not fully utilized
Solution Approach 1:
The patent substitutes digital signal processing with optical domain processing. By implementing compressed sensing in the optical domain using optical modulators and optical detectors, the system enables true concurrent processing where compression and recognition occur simultaneously during optical image acquisition, fully utilizing parallel processing capabilities rather than sequential digital processing.
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
This approach significantly reduces data storage and transmission requirements, enhances real-time performance, and facilitates concurrent processing, effectively addressing the limitations of prior art by implementing image recognition during imaging, thereby improving recognition efficiency.
Implementation Method 1
the filter unit controls the attenuation of the light intensity of the laser light, filters off high frequency scattered light, and forms a zero-order diffraction spot
Implementation Method 2
forms a zero-order diffraction spot, and then controls the diffraction spot to form parallel light
Implementation Method 3
the image generation unit generates an experimental image in which an object image and a specific measurement matrix are superimposed
Implementation Method 4
the image acquisition unit comprises a convergent lens and a single photon detector that are sequentially arranged in parallel with an optical path
Data Source
AI summary
A compressed sensing based object imaging system and an imaging method thereof. The object imaging system comprises a light source generation unit (11), a filter unit (12), an image generation unit (13), an image acquisition unit (14), and an image reconstruction unit (15). The light source generation unit (11) generates experimental laser; the filter unit (12) filters high frequency scattered light and forms parallel light; the image generation unit (13) generates an experimental image in which an object image (16) and a specific measurement matrix (17) are superimposed; the image acquisition unit (14) performs compression sampling on the generated experimental image; and the image reconstruction unit (15) reconstructs sampling data to restore the object image (16). The imaging method comprises: establishing a sample database comprising the specific target object image (16); training sample images to obtain the specific measurement matrix (17); and simultaneously completing image sampling, image compression and image recognition in an all-optical system. The system and the method can greatly reduce the data volume recorded in image recognition and image matching, thus improving the real-time performance of the system, and providing a possibility of concurrent processing by machine vision and artificial intelligence.