WiMi Developed A Single-pixel Digital Holography-based Compressive Phase Object Classification Technology

WiMi Hologram Cloud Inc., a leading global Hologram Augmented Reality (AR) Technology provider, announced the development of a single-pixel digital holography-based compressive phase target classification technology. It is a digital holography-based target identification and classification technology. The technology utilizes the principle of single-pixel digital holography imaging to capture the interferogram of a target by synthesizing the wave field in the object plane. And the interferogram is mapped to the CCD camera through the spatial light modulator for digital processing to extract the target phase information and then complete the target identification and classification process.

The core of this target classification technology is digital holography and phase extraction. The techniques it utilizes include:

Single-pixel digital holographic imaging principle: a laser is emitted to an object to produce an interferogram in the object plane, which records the wavefront information. The amplitude and phase information of the interferogram is recorded by calculating the complex amplitude and phase information and mapping them onto the single-pixel detector and the image diode array, respectively. Finally, synthetic wavefield and phase extraction techniques achieve the target classification and identification.

AiThority: How to Get Started with Prompt Engineering in Generative AI Projects

Image compression algorithm: The input interferometric image is compressed based on the AWP technique. AWP creates a multi-scale decomposition when compressing the image and extracts several local features of different sizes. These features compress the information of the interferometric picture into the smallest possible number of bits while retaining enough information to achieve target classification and recognition.

Phase extraction algorithm: With the help of the Fourier transform technique, the phase information of single-pixel digital holography is calculated based on a mathematical model. Since digital holographic imaging is performed in the frequency domain, the information in the frequency domain can be obtained after using Fourier transform. The phase information of the object is received after the phase extraction algorithm, which is the key to achieving target recognition and classification.

The single-pixel digital holography-based compressive phase target classification technique is highly compressible compared to traditional digital holography. Its use of image compression algorithms significantly reduces the number of pixels, thus greatly reducing the time and bandwidth required for transmission and improving data transmission efficiency. In addition, the technique uses the AWP technique to extract multiple local features of different sizes, which can improve the recognition rate and classification ability of targets and increase the image resolution.

Read: How AI NFTs Are Unlocking the Democratization of the Digital Economy

With the above technologies, WiMi‘s target classification technology can achieve fast, efficient, and accurate identification and classification of targets, with the advantages of high data transmission efficiency and low cost can be widely used in different scenarios.

The single-pixel digital holography-based compressive phase target classification technology has a wide range of application scenarios. For example, in the field of urban management, this technology can realize the monitoring of urban public facilities and land use and classify and identify buildings, roads, gardens, parks, etc., in cities. This is important for urban planning and management, enabling more accurate land use and building management. In the medical field, the technology can be applied to medical imaging to achieve high-precision imaging of the human body. For example, imaging the oral cavity, ophthalmology, skin, and other body parts is suitable for imaging different tissues in various forms. In industrial inspection, the technology can be applied to automated assembly, precision inspection of mechanical components, etc., to achieve high-precision imaging and inspection of parts. In system security and network protection, the technology can be used as an auxiliary technology for facial recognition and fingerprint recognition to achieve more accurate and efficient identity authentication and guarantee system security and network protection.

Latest Insights: Why Only AI and Data Analytics Can Stop Financial Criminals

[To share your insights with us, please write to sghosh@martechseries.com]