Abstract: Embedded intelligent vision sensor technology has become a research hotspot of wireless vision sensor network (WVSN) due to its low cost and high efficiency image capturing. In earlier research, a low cost and high resolution agricultural vision sensor (HRAVS) was developed based on the platform S3C6410 and OV3640 in South China Agricultural University. In this paper, a vision sensor remote transmission control schema (VSRTC) was developed to enable HRAVS to communicate with each other in various communication technologies (cable, 3G, 4G and Wi-Fi). The combination of HRAVS and VSRTC can be applied in many areas of the Internet of Things (IOT) in Agriculture. This paper introduced design of application architecture, transmission control protocol, and the node's application software of the VSRTC-HRAVS. A WVSN test was conducted for 25 days with 10 camera nodes in experimental fields of South China Agricultural University. Node control stability, the image capturing and encoding performance, the overall average image capturing time and the average frame rate of video capturing under different resolutions were evaluated in a series tests. The results showed that the new camera nodes were able to effectively carry out 3 capture modes (command response/cycle response/video), and under the re-transmission mode, the instruction loss rate was below 1% of all nodes. Given image pixel of 1.3, 2.0 and 3.2 Mpixel, when running without networking, the shortest average overall image processing time of the node were 6.2, 8.2 and 11.1 s respectively, and the largest video frame rates were 58.7, 34.6 and 16.4 frames per second, respectively; When running networking, the shortest average overall image processing time of the node were 17.6, 26.9 and 49.6 s respectively, and the highest video frame rates were 20.2, 16.1 and 9.3 frames per second, respectively. This scheme supported high resolution image and video transmission which can be applied in the field of agriculture where real-time transmission is not highly demanded.