The KITSUNE satellite is a 6-unit CubeSat platform with the main mission of 5-m-class Earth observation in low Earth orbit (LEO), and the payload is developed with a 31.4 MP commercial off-the-shelf sensor, customized optics, and a camera controller board. Even though the payload is designed for Earth observation and to capture man-made patterns on the ground as the main mission, a secondary mission is planned for the classification of wildfire images by the convolution neural network (CNN) approach. Therefore, KITSUNE will be the first CubeSat to employ CNN to classify wildfire images in LEO. In this study, a deep-learning approach is utilized onboard the satellite in order to reduce the downlink data by pre-processing instead of the traditional method of performing the image processing at the ground station. The pre-trained CNN models generated in Colab are saved in RPi CM3+, in which, an uplink command will execute the image classification algorithm and append the results on the captured image data. The on-ground testing indicated that it could achieve an overall accuracy of 98% and an F1 score of a 97% success rate in classifying the wildfire events running on the satellite system using the MiniVGGNet network. Meanwhile, the LeNet and ShallowNet models were also compared and implemented on the CubeSat with 95% and 92% F1 scores, respectively. Overall, this study demonstrated the capability of small satellites to perform CNN onboard in orbit. Finally, the KITSUNE satellite is deployed from ISS on March 2022.

Earth observation (EO) missions remain a challenging task for small satellite platforms due to their demanding requirements and space environment effects. In this study, the camera payload development and mission requirements are presented together with the ground-based testing results for a 6U CubeSat called KITSUNE, operating at low Earth orbit. The major challenge of the payload development is maintaining the focus of the optical system despite the thermal vacuum environment in orbit since the low thermal capacity and rapid temperature variation of CubeSats hinder the camera focus. First, the payload is developed with an objective of a 5-m-class imaging mission, which has a 31.4 MP CMOS sensor and a lens with a 300-mm focal length. Second, polyimide heaters and multilayer insulators are utilized in order to maintain focus during imaging operations. Third, a collimator lens is used to aid in image capture during thermal vacuum tests. These images are analyzed thoroughly using various focus measure operators. The Diagonal Laplacian was found to be the most suitable operator due to the consistency in test results. The results also showed that the heat generated by the camera sensor significantly affects the lens temperature and, ultimately, the target temperature of the lens was defined at −1.8°C. Finally, the test results are discussed, including thermal vacuum, vibration, total ionization dose, and the effect of exposure to direct sunlight on the CMOS sensor.