With the widespread adoption of Internet of Things (IoT) technology, an increasing number of traffic monitoring sensors are being deployed on urban roads to provide real-time intelligent monitoring services for traffic management. To address the characteristics of intelligent transportation systems with high real-time requirements and large data volumes, this paper proposes a real-time traffic data processing platform based on Flink that provides high throughput and low latency. The platform dynamically accesses Kafka distributed message queues to ingest real-time traffic monitoring source data, creates a visual monitoring environment, and evaluates the real-time computing performance of Flink. Experimental results show that when the number of Kafka partitions is 3, the access throughput reaches 51.46 MB/s. Flink achieves millisecond-level real-time computation under different parallelism degrees, enhancing the efficiency of real-time data processing. Additionally, the query times of the Partial Bloom Filter algorithm are effectively reduced by 86%. Therefore, our proposed system demonstrates superior computing performance and lower latency, meeting the stringent requirements of intelligent transportation scenarios and providing a solution for improving intelligent transportation application services.