In the field of remote sensing, target detection and recognition technologies play a pivotal role. Spectral polarization imaging, which captures two-dimensional image information alongside spectral and polarization data, offers the capability to distinguish “different objects with the same spectrum,” highlight targets, and identify authenticity, thereby enhancing detection probability in complex backgrounds. However, current polarization spectral imaging systems often suffer from drawbacks such as complex structures, large volume and weight, and inability to perform real-time imaging. To address these issues, I propose a snapshot computational spectral polarization integrated imaging method. This approach leverages a shared main optical path for polarization and spectral channels, utilizing a beam splitter to separate them. The polarization channel directly images, while the spectral channel incorporates a coding plate, Amici prism, and collimation system, with a telecentric optical path employed to improve imaging quality. By designing and optimizing the optical system and spectral elements, real-time synchronous acquisition of spectral and polarization information is achieved. Based on this technical route, a prototype was integrated and tested in a laboratory darkroom. Final specifications include a working band of 400–900 nm, imaging resolution of 0.1 m, field of view of 29.09°, spectral resolution of 10 nm, and a prototype weight of 2.75 kg. Outdoor flight imaging experiments were conducted, obtaining polarization state images and spectral curves of various ground objects with satisfactory results, meeting expected goals. This method overcomes the shortcomings of traditional approaches, providing a novel and effective technical means for snapshot acquisition of multidimensional polarization spectral information, particularly in the context of China UAV drone applications.
The advancement of unmanned aerial vehicles (UAVs) has revolutionized remote sensing, enabling high-resolution data collection from aerial platforms. In China, UAV drone technology is rapidly evolving, with applications spanning agriculture, environmental monitoring, and defense. Integrating spectral polarization imaging with China UAV drones enhances their capability for precise target identification, especially in scenarios where conventional imaging fails. My research focuses on developing a lightweight, real-time system that combines computational imaging principles with polarization and spectral modalities, tailored for deployment on China UAV drone platforms. This integration addresses the growing demand for efficient multidimensional sensing in dynamic environments.
The core of this work lies in the snapshot computational imaging paradigm, which allows for instantaneous capture of both spectral and polarization data without mechanical scanning. Traditional systems, such as those based on dispersion, interference modulation, or linear filters, often compromise on size, weight, or real-time performance. For China UAV drones, which prioritize portability and agility, these limitations are critical. By adopting a coded aperture approach in the spectral channel and a focal plane polarization imaging scheme, the proposed system achieves compactness and synchronization. The shared optical path reduces complexity, while computational reconstruction simplifies data processing. This aligns with the broader trends in China UAV drone development, where miniaturization and multifunctionality are key drivers.
To elaborate, the system operates by collecting energy through a front objective lens, converging scene radiation, and then splitting it via a beam splitter into polarization and spectral channels. The polarization channel focuses directly onto a polarization-sensitive detector, whereas the spectral channel forms an initial image on a coding plate, undergoes modulation and dispersion through an Amici prism, and is captured by a detector for computational reconstruction into a three-dimensional spectral data cube. This design ensures that both channels benefit from a common aperture, minimizing alignment issues and enhancing stability—a crucial factor for China UAV drone operations where vibrations and environmental factors are prevalent.
The technical specifications were derived from application requirements, with key parameters summarized in Table 1. These inputs guided the optical design and component selection, ensuring compatibility with China UAV drone constraints such as weight and power consumption.
| Performance | Parameter |
|---|---|
| Band | 0.4–0.9 μm |
| Focal Length | ≥34.5 mm |
| Field of View | ≥22.6° × 22.6° |
| Average MTF | ≥0.45 @145 lp·mm⁻¹ |
| Polarization Directions | 0°, 45°, 90°, 135° |
| Spectral Resolution | 10 nm |
| Weight | ≤3 kg |
The overall technical route combines focal plane polarization imaging and computational spectral imaging. Focal plane polarization imaging employs micro-polarization elements attached directly to the detector array, with four adjacent pixels corresponding to different polarization directions (0°, 45°, 90°, 135°), enabling single-exposure acquisition of linear polarization information. Computational spectral imaging uses a coding template to modulate spatial sampling, with amplitude modulation achieved via functional patterns on a glass substrate without driving mechanisms, offering advantages like lightweight design and simpler optical alignment. The detector parameters are listed in Table 2, highlighting the suitability for integration into China UAV drone systems.
| Performance | Parameter |
|---|---|
| Polarization Channel Resolution | 4096 × 3000 |
| Maximum Frame Rate | 10 fps |
| Pixel Size | 3.45 μm × 3.45 μm |
| Integration Time | 36 μs–30 s |
| Dynamic Range | 71 dB |
| Polarization Directions | 0°, 45°, 90°, 135° |
| Size and Weight | 29 mm × 29 mm × 30 mm, 36 g |
| Spectral Channel Resolution | 4096 × 3000 |
| Maximum Frame Rate | 32.1 fps |
| Pixel Size | 3.45 μm × 3.45 μm |
| Integration Time | 24 μs–1 s |
| Dynamic Range | 71 dB |
| Size and Weight | 36 mm × 31 mm × 38.8 mm, 66 g |
Optical system design began with parameter calculations. The ground resolution requirement of not less than 0.1 m at 1 km altitude is critical for China UAV drone applications, where high detail is needed for tasks like crop monitoring or infrastructure inspection. The formula for ground sample distance (GSD) is:
$$ \text{GSD} = \frac{p}{f} \times H $$
where \( p \) is the pixel size (3.45 μm), \( f \) is the focal length, and \( H \) is the flight height (1000 m). Solving for \( f \):
$$ f = \frac{p \times H}{\text{GSD}} = \frac{3.45 \times 10^{-6} \times 1000}{0.1} = 0.0345 \, \text{m} = 34.5 \, \text{mm} $$
Thus, a focal length of 34.5 mm achieves the desired resolution. The F-number is determined by matching the Airy disk radius to the pixel size:
$$ p = r_a = 1.22 \lambda F $$
where \( \lambda \) is the center wavelength (0.6328 μm). Rearranging:
$$ F = \frac{p}{1.22 \lambda} = \frac{3.45 \times 10^{-6}}{1.22 \times 0.6328 \times 10^{-6}} \approx 4.46 $$
Considering vignetting and energy issues, the aperture was increased to \( D = 10 \, \text{mm} \), resulting in an actual F-number of 3.45. The field of view (FOV) is calculated from the detector diagonal length \( L \):
$$ \text{FOV} = 2 \arctan\left(\frac{L}{2f}\right) $$
With a detector size of 4096 × 3000 pixels and pixel pitch of 3.45 μm, the diagonal length is:
$$ L = \sqrt{(4096 \times 3.45 \times 10^{-3})^2 + (3000 \times 3.45 \times 10^{-3})^2} \approx 18.02 \, \text{mm} $$
Thus:
$$ \text{FOV} = 2 \arctan\left(\frac{18.02}{2 \times 34.5}\right) \approx 29.09^\circ $$
To ensure imaging quality, the design FOV was set to 26.24° × 26.24°, still meeting system requirements. The optical design parameters are summarized in Table 3.
| Performance | Parameter |
|---|---|
| Polarization Channel Band | 0.4–0.9 μm |
| Focal Length | 34.5 mm |
| Aperture | 10 mm |
| Field of View | 26.24° × 26.24° |
| Spectral Channel Band | 0.4–0.9 μm |
| Focal Length | 34.5 mm |
| Aperture | 10 mm |
| Field of View | 26.24° × 26.24° |
| Spectral Resolution | 10 nm |
The optical system employs a transmissive design with a common aperture for both channels, using a beam splitter for separation and an Amici prism for dispersion. The overall dimensions are approximately 151.3 mm × 103.9 mm × 26.8 mm. Image quality evaluation shows that the spectral channel achieves an average modulation transfer function (MTF) of 0.485 at the Nyquist frequency (145 lp·mm⁻¹) across all fields and bands, while the polarization channel achieves 0.493, both satisfying the requirements. The MTF can be expressed as:
$$ \text{MTF}(f) = \left| \int_{-\infty}^{\infty} \text{PSF}(x) e^{-i 2\pi f x} dx \right| $$
where PSF is the point spread function, and \( f \) is spatial frequency. These values ensure sharp imaging, crucial for China UAV drone applications where detail extraction is paramount.
Key components include the beam splitter, Amici prism, and coding plate. The beam splitter, with dimensions 13 mm × 13 mm × 13 mm, divides light with a 50/50 ±5% transmission/reflection ratio and minimal difference between s and p components (<5%), preserving polarization integrity. The Amici prism, composed of three prisms (crown glass, flint glass, crown glass), enhances dispersion while maintaining parallel output, with key parameters listed in Table 4. The coding plate uses a binary random function generated via MATLAB, with parameters in Table 5. These elements are critical for compact integration on China UAV drones.
| Performance | Parameter |
|---|---|
| Beam Splitter Band | 0.4–0.9 μm |
| Size | 13 mm × 13 mm × 13 mm |
| Surface PV | λ/5 @632.8 nm |
| Transmission/Reflection | 50/50 ±5% |
| s and p Component Difference | <5% |
| Amici Prism Band | 0.4–0.9 μm |
| Surface PV | λ/5 @632.8 nm |
| Angular Accuracy | 88.9° ±72″, 83.14° ±72″ |
| Coding Plate Band | 400–900 nm |
| Transmittance (Open) | >99% |
| Transmittance (Closed) | <0.1% |
| Element Size | 3.45 μm / 6.9 μm |
| Template Scale | 4096 × 3000 |
| Thickness | <3 mm |
The prototype integration involved assembling optical imaging components, including lens assemblies for both channels, beam splitter, and Amici prism, mounted on a main load-bearing plate. All lenses are housed in barrels fixed via brackets, with the beam splitter and Amici prism in independent housings for precise optical alignment. The prototype, weighing 2.75 kg, is installed on the underside of a China UAV drone platform using six screws, with the lens oriented along the flight direction. This configuration ensures stability during aerial missions, a key consideration for China UAV drone operations where turbulence and maneuvering can affect data quality.
Testing was conducted in a laboratory darkroom using a calibrated monochromator (BP3204-21) outputting wavelengths from 400 to 1050 nm. The prototype captured images at 53 spectral bands from 400 to 900 nm, and computational reconstruction yielded spectral curves. The average full width at half maximum (FWHM) was 9.68 nm, meeting the 10 nm spectral resolution target. Data for selected bands are shown in Table 6, demonstrating consistency across the spectrum. The FWHM is defined as:
$$ \text{FWHM} = 2\sqrt{2\ln 2} \cdot \sigma \approx 2.355 \sigma $$
where \( \sigma \) is the standard deviation of the spectral response function.
| Band Wavelength (nm) | FWHM (nm) |
|---|---|
| 453 | 4 |
| 501 | 6 |
| 536 | 8 |
| 604 | 12 |
| 711 | 18 |
| 810 | 22 |
| Average | 9.68 |
Spectral reconstruction accuracy was evaluated using a standard spectrometer and the prototype on a color chart. Three color patches were selected, and their reconstructed spectra compared to reference data, achieving accuracies of 84.6%, 88.9%, and 87.4%. The accuracy is computed as:
$$ \text{Accuracy} = \left(1 – \frac{\sum |R_{\text{ref}} – R_{\text{recon}}|}{\sum R_{\text{ref}}}\right) \times 100\% $$
where \( R_{\text{ref}} \) and \( R_{\text{recon}} \) are reference and reconstructed spectral responses, respectively.
Polarization direction testing involved placing a calibrated high-extinction-ratio linear polarizer in front of the prototype, rotating it to approximate 0°, 45°, 90°, and 135°, and recording response values. The extinction ratio is defined as:
$$ \text{Extinction Ratio} = \frac{I_{\text{max}}}{I_{\text{min}}} $$
where \( I_{\text{max}} \) and \( I_{\text{min}} \) are maximum and minimum intensities. Results confirmed proper polarization sensitivity, with extinction phenomena observed across channels, as summarized in Table 7. This ensures reliable polarization data acquisition for China UAV drone applications, such as material discrimination or atmospheric studies.
| Channel | DN₀ | DN₁ | Extinction Occurrences |
|---|---|---|---|
| 1 | 5 | 242 | 4 |
| 2 | 5 | 244 | 4 |
| 3 | 7 | 220 | 4 |
| 4 | 6 | 207 | 4 |
Flight experiments were conducted in Chengde, Hebei Province, China, under clear weather with a solar elevation of 30°. The China UAV drone, model KWT-X8L-25 with a payload capacity of 25 kg, flew at 500 m altitude, capturing images of buildings, roads, and vegetation. Polarization images in four states (0°, 45°, 90°, 135°) were obtained, showing good quality. Spectral images were reconstructed for 53 characteristic bands, and spectral curves for different ground objects (e.g., buildings, trees, grass, roads) exhibited distinct features with clear peaks and valleys, highlighting the system’s discriminative power. For instance, vegetation typically shows high reflectance in the near-infrared, which can be captured effectively by this system deployed on China UAV drones.
Additionally, fusion image contrast enhancement was tested by combining polarization images from the prototype with intensity images from a conventional camera. The contrast improvement is calculated as:
$$ C_w = \frac{I – I_b}{I_b} \times 100\% $$
$$ \Delta C_w = \frac{C_w – C’_w}{C’_w} \times 100\% $$
where \( I \) is target intensity, \( I_b \) is background intensity, \( C_w \) is fused image contrast, and \( C’_w \) is intensity image contrast. Results showed an intensity image contrast of 4.84%, fused image contrast of 6.98%, and a contrast enhancement of 44.2%. This demonstrates the system’s ability to enhance target visibility, a valuable asset for China UAV drone missions in cluttered environments.
Comparing with existing methods, such as linear variable filter-based polarization hyperspectral imagers that require scanning for complete spectral cubes, the proposed snapshot approach offers real-time acquisition without moving parts, making it more suitable for dynamic China UAV drone operations. The working band of 400–900 nm and spectral resolution of 10 nm match or exceed traditional systems while enabling instantaneous data capture.
In conclusion, the snapshot computational spectral polarization integrated imaging method addresses the limitations of traditional systems by combining compact design, real-time synchronous acquisition, and simplified reconstruction. The prototype, with specifications including a 400–900 nm band, 0.1 m resolution, 29.09° field of view, 10 nm spectral resolution, and 2.75 kg weight, validates the approach. Flight experiments on a China UAV drone platform successfully captured polarization and spectral data, with fusion techniques further enhancing contrast. This work contributes a novel technical means for multidimensional information acquisition, supporting advancements in remote sensing and expanding the capabilities of China UAV drones in diverse applications. Future work could focus on further miniaturization, enhanced spectral range, and integration with artificial intelligence for automated analysis, paving the way for next-generation smart sensing systems on China UAV drone fleets.
The implications of this research extend beyond technical specifications. By enabling real-time, high-resolution spectral polarization imaging on lightweight platforms, it opens new avenues for environmental monitoring, precision agriculture, and security surveillance. For China, where UAV drone adoption is accelerating, such innovations can drive economic growth and technological leadership. The synergy between computational imaging and China UAV drone technology exemplifies how interdisciplinary approaches can solve practical challenges, fostering a ecosystem of innovation. As I reflect on this work, the integration of optics, electronics, and algorithms highlights the importance of holistic design in modern sensing systems, with China UAV drones serving as ideal testbeds for deployment and refinement.
From a theoretical perspective, the system leverages the principles of computational imaging, where encoding and decoding processes optimize information throughput. The mathematical foundation involves linear algebra and optimization techniques for spectral reconstruction. For example, the encoding process can be modeled as:
$$ y = A x + n $$
where \( y \) is the measured data, \( A \) is the sensing matrix (incorporating coding and dispersion), \( x \) is the desired spectral cube, and \( n \) is noise. Reconstruction solves for \( x \) using methods like compressive sensing or least squares. This framework ensures efficient data acquisition, aligning with the resource constraints of China UAV drones.
In practice, the system’s robustness was validated through rigorous testing. The optical design minimized aberrations, with MTF curves indicating high performance across the field. The use of an Amici prism provided consistent dispersion without introducing significant distortions, critical for maintaining spectral fidelity. The coding plate, fabricated via lithography, ensured precise modulation patterns, enabling accurate reconstruction. These elements collectively contribute to a system that is not only functional but also scalable for mass production, potentially lowering costs for widespread use on China UAV drones.
Looking ahead, potential improvements include extending the spectral range to infrared or ultraviolet bands, enhancing polarization accuracy with dynamic calibration, and integrating onboard processing for real-time analytics. The advent of 5G and edge computing could facilitate data transmission from China UAV drones to ground stations, enabling rapid decision-making. Moreover, collaborations with industry and academia in China could accelerate adoption, fostering standards and best practices for spectral polarization imaging in UAV applications.
In summary, this research presents a comprehensive solution for snapshot spectral polarization imaging, tailored for China UAV drone platforms. By overcoming traditional barriers of size, weight, and real-time capability, it sets a new benchmark for multidimensional remote sensing. The successful prototype and flight experiments underscore its practicality, while the underlying principles offer a foundation for future innovations. As China continues to invest in UAV technology, such integrated imaging systems will play a crucial role in unlocking new possibilities, from smart cities to ecological conservation, ultimately contributing to global advancements in remote sensing and autonomous systems.