The invention discloses a code rate control method for near lossless compression of a JPEG-LS (Joint Photographic Experts Group-Lossless) image with near optimal rate distortion performance. The code rate control method for near lossless compression is designed specific to a JPEG-LS standard. A mathematical model between an average absolute predicted difference and a code quantization parameter NEAR relevant to an encoding rate at a high code rate is pre-established. In a process of encoding a specific image, a quantization parameter is adjusted by the model in a way of taking a block as a unit. According to adjustment logic, the sum of the number of bits consumed by an encoded part and the estimated number of bits required by a non-encoded part under a preset quantization parameter is closet to a predetermined code rate requirement. Through adoption of the method, large-scale fluctuation of the quantization parameter can be avoided as much as possible, thereby realizing near optimal rate distortion performance.