This paper presents a newly developed architecture for a highly parallel DSP suited for realtime image reaognition. The programmable DSP was designed for a variety of image recognition systems, such as computer vision systems, character recognition systems and others. The DSP consists of functional units suited for image recognition: a SIMD processing core, a hierarchical bus, an Address Generation Unit, Data Memories, a DMA controller, a Link Unit, and a Control Unit. The high performance of 3.2GOPS is realized by the eight-parallel SIMD core with a optimized pipeline structure for image recognition algorithms. The DSP supports flexible data transfers including an extraction of lacal images from raster scanned image data, a table-loop-up, a data-broadcasting, and a data-shifting among processing units in the SIMD core, for effective execution of various image processing algorithms. Hence, the DSP can process a 55 spatial filtering for 512512 images within 13.1 msec. Adopting the DSP to a Japanese character recognition system, the speed of 924 characters/sec can be achieved for feature extractions and feature vectors matchings. The DSP can be integrated in a 14.514.5 mm2 single-chip, using 0.5 µm CMOS technology. In this paper, the key features of the architecture and the new techniques enabling efficient operation of the eight parallel processing units are described. Estimation of the performance of the DSP is also presented.