Camouflaged object detectionCamouflaged object detection (COD) is critical in understanding the relationship between animals and their surrounding environments. It also finds applications on battlefields since enemy objects (humans and weapons) tend to be hidden from view. State-of-the-art COD methods, which rely on deep neural networks, are computationally demanding and exhibit marginal performance variations across different models. As important, their effectiveness is limited by the size of training samples. Whenever camouflaged scenes deviate from the training data, their detection performance degrades. To address these challenges, we propose a new camouflaged object detectionCamouflaged object detection paradigm that exploits augmented intelligenceAugmented intelligence through humanHuman-AI-AIArtificial Intelligence (AI) collaboration. Within this paradigm, humans and AIArtificial Intelligence (AI) examine the same scene and identify complementary camouflaged objects. AIArtificial Intelligence (AI) typically detects camouflaged objects seen before, while humans excel at detecting new camouflaged objects that AIArtificial Intelligence (AI) usually misses due to the lack of sufficiently diverse training samples. Once camouflaged objects are detected, they can be tracked in subsequent frames. Simply stated, augmented intelligenceAugmented intelligence is a more robust approach than deep learningDeep learning used in isolation. This paper presents a novel augmented intelligenceAugmented intelligence solution called GreenCOD, which leverages gradient boosting and deep learningDeep learning features extracted from a pre-trained EfficientNet. It significantly simplifies model design, resulting in a system that requires fewer parameters and operations and maintains high performance compared to state-of-the-art deep neural network models. Our AIArtificial Intelligence (AI) models, which are trained without backpropagation, achieve superior performance with fewer than 20G Multiply-Accumulate Operations. This more efficient paradigm opens new avenues for further exploration of augmented intelligenceAugmented intelligence and environmentally friendly AIArtificial Intelligence (AI) techniques.