Abstract: Abstract: Traditional single monitoring technique in orchard environment has such shortages as weak effectiveness, inaccurate count and pooruniversality. Now existing pest monitoring methods include acoustic measurement, piezoelectric measurement, infrared measurement and machine vision recognition technology. In view of this, the future development trend of pest detection technology will undoubtedly be a variety of detection methods combined with each other. Comprehensive utilization of the existing testing methods will form a multiple information fusion technique to detect and provide reliable scientific decision based on comprehensive prevention and control of fruit pests, and the loss will be reduced to a minimum. In this paper, infrared measurement and machine vision recognition technology are integrated to identify pest species and count pest populations, and information of pests is obtained from 2 aspects. The accuracy of the fusion result is verified by comparing with the manual count. Taking Grapholitha molesta, Dichocrocis punctiferalis, Adoxophyes orana and disruptors as research objects, recognition results of infrared sensors and machine vision are obtained using the laboratory artificially randomly scattered test samples. Test samples were collected in Xiaotangshan National Precision Agriculture Research and Demonstration Base from July to September in 2015. For the infrared method, infrared circuit is mainly composed of infrared detector, photoelectric detector, filter, amplifier, communication module, and so on. Due to the different size of insect pests, the infrared output is different. The bigger the pest, the bigger the value of the infrared output. Therefore, the influence of ambient light on the detection results is significant. For example, Adoxophyes orana is larger than Grapholitha molesta and smaller than Dichocrocis punctiferalis. To go along with this, the thresholds of Grapholitha molesta, Adoxophyes orana, Dichocrocis punctiferalis and disruptors are 5.655, 13.47 and 23.13, respectively. The system is mainly composed of infrared sensor unit and machine vision unit. The infrared sensor unit introduces the phase lock amplifier technology to extract the weak useful signal from the noise environment, and to solve the problem of the influence of the natural light environment. The core of the lock-in amplification technology is correlation detection, and using the characteristic of useful signals and noise signals being not related to each other to extract the useful signal from the noise by the correlation operation. Using Matlab environment feature extraction algorithm, normalized entropy and normalized energy are chosen as texture feature indices for the HSV three-channel texture feature based on the 'DB4' wavelet decomposition. Infrared image fusion and pest identification are mainly based on the time stamp of infrared and image recognition. Fusion count results are obtained by a formula operation which is derived from the linear regression analysis of SPSS. The results of infrared sensor and machine vision are the input of the formula. We can get the conclusion that the infrared output value ranges are (0,5, (5,13, (13,23, and (23,32, and the infrared recognition accuracy rates of Grapholitha molesta, Dichocrocis punctiferalis, Adoxophyes orana and disruptors are 92%, 78%, 80% and 88% respectively. The image recognition accuracy rates are 92%, 88%, 92% and 90%, respectively, and the fusion recognition accuracy rates are 98%, 92%, 94% and 96%, respectively. Obviously, the fusion of infrared sensor and image recognition technology can improve the accuracy and efficiency of the identification of fruit pests. This method has very high innovation in both theory and practical application.