The insect pests and crop diseases are the most critical factors that affect agricultural production, which reduces the sustainable development of agriculture. While detecting the pest, it is inconsistent to place the surveillance cameras near the target pests and the captured images from the Internet of Things (IoT) monitoring equipment at a constant location that is mostly insufficient for pest detection. IoT is a well-known advanced technology and an analytics system incorporated in diverse industries based on its unique abilities and flexibilities over a particular environment like agriculture. There is a demand for the IoT in agricultural areas to reduce the chemical crop protection agents and fertilizers to manage the efficient crop state and crop production. Hence, the data collection is through the IoT devices in this research model. This research aims to develop a pest identification and classification model to detect and identify the pests in the images. Initially, the IoT platform is created, and IoT devices conduct the data collection. Then, object detection is performed using Yolov3 to detect the pests in the images from the gathered images. The detected images are subjected to the Convolutional Neural Network (CNN) for gathering the deep features, which are then forwarded to the enhanced classifier, termed Convolution Neural Long Short-Term Memory (CNLSTM) for getting the classified outcomes as pest details, in which the optimization of parameters is done by Adaptive Honey Badger Algorithm (AHBA). These results demonstrated that the proposed method shows enhanced performance by rapidly collecting the information in agriculture and ensures the technical indication for population estimation and pest monitoring.