Secure image transmission using chaotic encryption and DWT watermarking on reconfigurable platform

With the exponential rise in digital image transmission across unsecured networks, ensuring image confidentiality and authenticity has become essential. Traditional cryptographic methods like AES provide strong confidentiality but lack built-in authentication mechanisms. On the other hand, digital watermarking, especially in the Discrete Wavelet Transform (DWT) domain, offers robust image authentication. Combining both techniques and deploying them on reconfigurable platforms such as FPGAs addresses the growing demand for secure, high-performance image communication systems. This paper presents a hybrid image authentication framework that integrates AES encryption for confidentiality and DWT-based invisible watermarking for authenticity. The process involves preprocessing the image, applying AES encryption, embedding a watermark in the high-frequency sub-bands of the DWT domain, and implementing the system on an FPGA using Verilog HDL. The AES and DWT modules are designed as separate cores, synthesized, and verified using Xilinx tools on Spartan-6 FPGA. The watermark, encrypted if necessary, is embedded in the HH sub-band to ensure robustness and imperceptibility. Simulations conducted on standard grayscale images (e.g., Lena, Cameraman) demonstrate high image fidelity, with PSNR exceeding 40 dB and SSIM values above 0.95. The watermark remained resilient under Gaussian noise, JPEG compression, and median filtering, with a bit error rate below 2%. FPGA implementation showed efficient resource utilization, consuming 4321 LUTs and 2176 flip-flops, with a total power consumption of 354 mW and latency of 1.02 μs. These results confirm the system’s real-time suitability and robustness. The proposed hybrid image authentication system successfully ensures both security and authenticity in an image.