Visible Light Communication (VLC) positioning technology holds great potential and application prospects in the field of wireless communication. To enhance the efficiency and broaden the application scope of VLC for indoor positioning, a balanced code encoding and decoding system for optical communication-based indoor positioning are proposed. The system first develops a light barcode encoding system based on the characteristics of LED lamps, leveraging the periodic characteristics of optical signals for data transmission. Subsequently, the advanced YOLOv8 algorithm is introduced to accurately locate the cyclic visible light codes emitted, enabling accurate capture of the coding period. Additionally, a decoding algorithm incorporating skew correction and error correction techniques is designed to optimize the decoding process of the located images, thereby improving data accuracy and stability. Experimental results show that the YOLOv8-based image recognition system can quickly and accurately identify and decode the encoded signals emitted by LED lamps, with an average recognition time of 0.324 seconds. The decoding efficiency is doubled compared to traditional methods. This approach significantly enhances and extends the performance of VLC-based indoor positioning systems, showing promising potential for future automation systems and contributing to technological advancements in related industries.