Softwarized Unmanned Aerial Vehicles (UAVs) use network programmability concept of Software-Defined Network (SDN) to separate the hardware control layer from the data layer via OpenFlow protocols. The softwarized UAV enable ubiquitous connection, as well as a flexible, cost-effective, and improved method for upgrading all network services without shutting down the entire system. However, the connectivity of UAVs with OpenFlow switches and their heavy reliance on unsecured communication protocols makes the entire network vulnerable. This is a critical concern, particularly in combat surveillance, where eavesdropping, adding, changing, or deleting messages during communications between deployed UAVs and SDN controller is a possible threat. To mitigate the aforementioned issues, this paper presents a novel secure data sharing framework for softwarized UAV environments that incorporates blockchain and Deep Learning (DL). First we present a blockchain-based technique to reg-ister, verify and thereafter validate the communication entities in softwarized UAV environment using smart contract-based Proof-of-Authentication (PoA) consensus mechanism. Additionally, a new deep neural network architecture-based flow analyzer is designed to detect illegitimate transactions. The latter combines a Stacked Contractive Sparse AutoEncoder with Attention-based Long Short-term Memory Neural Network (SCSAE-ALSTM) to improve intrusion detection process. The effectiveness of our framework over several standard baseline methodologies is demonstrated by security analysis and experimental findings. © 2022 IEEE.