This study evaluates the process of detecting the degree of firmness and pH of rose apples with the help of image processing from the point of view of environmental effects. The process of this study started with image processing. In image processing, the selected samples were photographed with a charge-coupled device (CCD) camera, and red (R), green (G), and blue (B) values were extracted with the image processing algorithm in MATLAB software. Next, the hardness and acidity values of the samples were extracted using laboratory steps. Next, with the inputs of each test, the life cycle assessment (LCA) list was prepared. Then, with the Impact 2002+ method, the list was subjected to life cycle evaluation, and the middle and final effects of the analyses were extracted. Next, the neural network and grey wolf optimizer (GWO) methods were used to predict environmental effects. Based on the results, it was determined that the values of R and G had the highest effect on estimating pH and the values of B and G had the highest effect on estimating the product's hardness. Also, the results of evaluating the accuracy of the artificial neural network combined with the grey wolf optimizer showed that the accuracy of the estimation of environmental effects in the evaluation of pH was about 3-5% higher than that of soluble solid content (SSC). Based on the findings, using the integrated machine learning system with image processing is a reliable method to estimate the environmental effects of detecting the quality characteristics of Iranian rose apples entirely non-destructively.