This thesis, titled ”Low Power Encoder for Embedded Monitoring,” focuses on the development of efficient image compression techniques tailored to the needs of embedded monitoring systems. These systems, often constrained by limited energy and computational resources, require innovative approaches to ensure effective data processing and transmission. The study empha-sizes low-complexity and energy-efficient algorithms utilizing the Discrete Cosine Transform (DCT) and Discrete Tchebichef Transform (DTT), which are widely used for data compression in digital media, telecommunications,and storage systems.A novel 8-point DCT approximation is proposed, achieving up to a 1 dB improvement in image quality compared to existing methods while main-taining a computational structure optimized for energy-sensitive embedded monitoring applications. A pruned version of this DCT further enhances the trade-off between performance and efficiency. Additionally, two new DTT approximations are introduced, reducing computational complexity and improving compression performance. These approximations are validated through FPGA implementations, demonstrating superior hardware perfor-mance and energy efficiency, with a quality gain of up to 2 dB compared to state-of-the-art DTT methods. Furthermore, the thesis presents a novel algorithm for generating cus- tomized quantization tables and coefficient orderings specifically designed for the proposed DCT and DTT approximations. This algorithm consis-tently improves image quality, delivering an average PSNR increase of 0.3 dB, thereby enhancing the overall performance of the encoder. The contributions of this research provide significant advancements in the design of low-power encoders for embedded monitoring systems. By addressing the challenges of energy efficiency and computational complexity, the proposed methods offer practical solutions for modern embedded systems and other resource-constrained signal processing applications.