Crop Identification via NDVI Time Series and Gaussian Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing crop identification methods face challenges due to spectral overlaps between different crops, leading to low classification accuracy, and are often specific to particular geographical areas and crops, making them difficult to apply universally.
Innovation Solution
A computing device-based method that uses multi-temporal remote sensing images and a multivariate Gaussian model to identify crops by calculating NDVI values, reducing noise, and constructing NDVI time series for accurate identification, without requiring professional agronomic knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If mono-temporal remote sensing images are used for crop identification, then the identification process is simple and fast, but classification accuracy is low due to spectral overlaps between different crops
Solution Approach 1:
The patent performs preliminary actions by calculating NDVI values for multiple temporal images before the actual identification. It pre-processes the remote sensing images by computing NDVI indices across different time points, constructing NDVI time series data that captures temporal variations in crop spectral characteristics. This preliminary preparation enables more accurate crop identification by leveraging temporal dynamics rather than relying on single-time-point spectral data alone.
2Adaptability or versatility
If classification rules specified by agricultural experts are used, then the identification can be accurate for specific crops in specific areas, but the rules are difficult to apply to different geographical areas and different crops
Solution Approach 1:
The patent achieves universality by developing a classification model based on NDVI time series characteristics that can be applied across different geographical areas and crop types. Instead of using region-specific expert rules, the model learns universal temporal spectral patterns through training data from multiple sources. The system can identify various crop types (cereals, legumes, oilseeds, etc.) across different locations by capturing their temporal NDVI signatures, making it a universally applicable tool without requiring location-specific recalibration.
3Measurement precision
If multi-temporal remote sensing images and NDVI time series are used for crop identification, then classification accuracy is improved by overcoming spectral overlaps, but the computational process becomes more complex
Solution Approach 1:
The patent extracts and emphasizes the most informative feature - the NDVI time series - from the complex multi-temporal remote sensing data. By transforming multiple spectral images across time into a single derived parameter (NDVI) calculated at each time point, and then analyzing the temporal sequence of these values, the system reduces the dimensionality of the data while retaining the critical temporal variation information needed for accurate crop identification. This extraction simplifies the computational process compared to analyzing all original spectral bands across all time points simultaneously.
Data Source
AI summary
In a crop identification method, multi-temporal sample remote sensing images labeled with first planting blocks of a specific crop are acquired. NDVI data of the sample remote sensing images are calculated. Noise of the NDVI data is reduced. A first multivariate Gaussian model is fitted based on de-noised NDVI data of the sample remote sensing image. Multi-temporal target remote sensing images are acquired. An NDVI time series of each pixel in the target remote sensing image is constructed. The NDVI time series is input to the first multivariate Gaussian model to obtain a likelihood value of each pixel displaying the specific crop in the remote sensing images. Second planting blocks of the specific crop in the target remote sensing images are determined accordingly. An accurate and robust identification result is thereby achieved.