Algorithm 1: Psuedocode of the proposed model

Input: Dataset samples S = {(X1, Y1),(X2, Y2),…,(Xn, Yn)}. The S is categorized into a training set (TrainX, TrainY), a validation Set (valx, Valy), and a testing set (testx, testy), where x is the number of pest images, and y is the corresponding image labels.

T denoted the number of training epochs.

Output: converge model

Load the (TrainX, TrainY), and (valx, Valy);

Augment the (TrainX, TrainY);

Begin:

Initialize weights and biases.

For m = 1, 2, 3, …, T:

Features extraction using CSP

Input the feature SK Attention Module

Generate the attention map using SK Attention Module

Fed the extraction features from SK Attention Module to Multiscale Feature Detection

Weight the multiscale feature maps, and calculate the output of the Multiscale Feature Detection.

Model fit (Optimizer, (TrainX, TrainY)) → (M(m))

Model evaluate (M(m), (ValX, ValY)) → mAP(m).

End For

Save the optimal model which has max mAP in T epochs.

End

Load the testing set;

Load the optimal model in terms of object detection performances.