We propose a novel architecture based on the strucuture of AutoEncoders. The paper introduces CrossEncoders - an AutoEncoder architecture which uses cross-connections to connect layers (both adjacent and non-adjacent) in the encoder and decoder side of the network respectively. The network incorporates both global and local information in the lower dimension code. We aim for an image compression algorithm that has reduced training time and better generalization property. The use of cross-connections makes the training of our network significantly faster. The performance of the proposed framework has been evaluated using real-world data from highly competitive datasets like MNIST and CIFAR-10. Furthermore, we show that the proposed architecture provides high compression ratio and is robust as compared to previously proposed architectures and PCA. The results were validated using metrics, such as PSNR-HVS and PSNR-HVS-M respectively.
This paper investigates the compression of infrared images with three codecs: JPEG2000, JPEG-XT and HEVC. Results are evaluated in terms of SNR, Mean Relative Squared Error (MRSE) and the HDR-VDP2 quality metric. JPEG2000 and HEVC perform fairy similar and better than JPEG-XT. JPEG2000 performs best for bits-per-pixel rates below 1.4 bpp, while HEVC obtains best performance in the range 1.4 to 6.5 bpp. The compression performance is also evaluated based on maximum errors. These results also show that HEVC can achieve a precision of 1°C with an average of 1.3 bpp.