Abstract: Golden cicada nymphs have been one of the most commonly edible insect species in the world, due to the extremely high medicinal and nutritional value with the unique taste. Therefore, the breeding of golden cicada nymphs can be brought the high economic value in food production. In this study, a rapid and accurate detection was proposed for the golden cicada nymphs in the natural environment. 366 images were captured from the Jiangsu University campus, Zhejiang Province, China, and another 180 images were from the green belt of residential area, and the downloaded from the Internet as supplement. Three procedures were proposed to realize the rapid and accurate detection of golden cicada nymphs, while reducing the model size and computational load. The first procedure was to reduce the number of branches of the original MobileNet-Single-Shot multibox Detection (SSD) network. Six SSD branches were used to detect the objects, each of which was corresponding to the different size of the feature map in original model. A relatively simple task was implemented to avoid the very complex neural network. Therefore, three small-scale branches of 3×3, 2×2, and 1×1 were removed to reduce the size of the model and the calculation amount. The second procedure was to reduce the width of MobileNet-SSD backbone, namely the input/output channels of the convolution layer. The number of features was determined in the network, where the enough width was used to ensure each layer learn the rich features. Among them, the width of the network was dominated the size of the parameters and calculation amount of network. A much wider network was used to extract much more the repetitive features, leading to the higher calculation amount without the substantial contribution to the network. Thus, the width of the backbone network was reduced significantly, due to only one type of object to identify in this case. The third procedure was to adjust the depth (the number) of the convolution layers. The depth of the neural network also determined the non-linear expression of the system. Generally speaking, the deeper neural networks presented the better nonlinear expression capabilities to learn more complex features. Thus, the depth of the convolution layer increased to improve the performance of the network. Three models were then proposed for the detection of golden cicada nymphs. The test results of night images show that the size and calculation amount of the improved model were greatly reduced with the normal network performance. Specifically, the size of the model was reduced from 15.22 to 1.51 MB, the floating-point operation amount was also reduced from 1.13×109 to 1.26×108. The average precision, the average Intersection over union (IoU), and the F1 score were 90.46%, 83.52%, and 92.35%, respectively, while, the GPU and CPU detection speed reached 179.3 and 6.49 frames/s, respectively. The daytime images show that the excellent generalization performance was achieved for the improved model. The improved MobileNet-SSD model presented the higher cost efficiency to greatly reduce the size of the model while reducing the computation load, while only a slight decrease in the performance, compared with the original. A comparative advantage was gained, compared with other lightweight target detection models. Therefore, the model can be widely expected to balance the performance, the size of the model and the computation load. The improved model can be very suitable for the deployment on mobile terminals and other embedded resource-constrained devices, which is conducive to real-time and accurate detection of golden cicada nymphs.