Performance of Interpolation Techniques for Compression of Crop Image: A Comparative Study

Image processing technology plays an important role in the field of cultivation. Its applications are rapidly increasing in agriculture fields such as disease identification, prevention, and crop health monitoring. Generally, good quality crop images are captured using high-resolution cameras, which need huge storage space and large bandwidth for transmission. Therefore, a strong motivation is there to develop an efficient algorithm for the compression and reconstruction of crop images. In this regard, the singular vector sparse reconstruction (SVSR) method is proposed by Xu et al., for image compression. This method is examined for compression and reconstruction of crop images using different types of interpolation methods such as Next-neighbor interpolation (NxNI, Previous-neighbor interpolation (PNI), Nearest-neighbor interpolation (NNI), Cubic spline interpolation (CSI), Piecewise Cubic Hermite Interpolating Polynomial (PCHIP), Makima interpolation (MI) and Linear interpolation (LI). The performance of these interpolation is assessed in terms of the following parameters such as compression ratio, peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and mean-square error (MSE). According to the results of the tests and comparisons, PCHIP and MI interpolation methods with SVSR technique has provided comparatively better performance for compression and reconstruction of crop images.