In this paper, a crop yield prediction system is proposed based on QR code and edge-cloud framework. In traditional systems, the cloud is used for IoT data analysis and storage. The concerns over the cloud-only system include security, latency, energy consumption, etc. The use of edge–cloud computing deals with the problem of latency and energy consumption. Nevertheless, the security is still a major concern. In the proposed system, the user uploads soil and environmental parameters’ data to the edge server using QR code to achieve secure data transmission. The data are analyzed using gated recurrent unit and long short-term memory network, and the result is sent to the cloud, from which the user can access the result anytime, anywhere. From the results, we observe that the proposed framework reduces the latency and energy consumption by \(\sim \) 28% than the cloud-only framework. The results also demonstrate that the proposed framework has an accuracy of \(\sim \) 99% in crop yield prediction.