Content Based Image Retrieval (CBIR) has gained a lot of interest over the last two decades. The need to search and retrieve images from databases, based on information (“features”) extracted from the image itself, is becoming increasingly important. CBIR can be useful for handheld image recognition devices in which the image to be recognized is acquired with a camera, and thus there is no additional metadata associated to it. However, most CBIR systems require large computations, preventing their use in handheld devices. In this PhD work, we have developed low-complexity algorithms for content based image retrieval in handheld devices for camera acquired images. Two novel algorithms, ‘Color Density Circular Crop’ (CDCC) and ‘DCT-Phase Match’ (DCTPM), to perform image retrieval along with a two-stage image retrieval algorithm that combines CDCC and DCTPM, to achieve the low complexity required in handheld devices are presented. The image recognition algorithms run on a handheld device over a large database with fast retrieval time besides having high accuracy, precision and robustness to environment variations. Three algorithms for Rotation, Scale, and Translation (RST) compensation for images were also developed in this PhD work to be used in conjunction with the two-stage image retrieval algorithm. The developed algorithms are implemented, using a commercial fixed-point Digital Signal Processor (DSP), into a device, called ‘PictoBar’, in the domain of Alternative and Augmentative Communication (AAC). The PictoBar is intended to be used in the field of electronic aid for disabled people, in areas like speech rehabilitation therapy, education etc. The PictoBar is able to recognize pictograms and pictures contained in a database. Once an image is found in the database, a corresponding associated speech message is played. A methodology for optimal implementation and systematic testing of the developed image retrieval algorithms on a fixed point DSP is also established as part of this PhD work.