<p>Afforestation is a practical way to absorb carbon dioxide in contemporary context of carbon neutrality. In the assessment of forest emission reduction projects, stock can be utilized as a crucial indicator of biomass and carbon storage estimation in addition to serving as an economic value index for forest wood resources. We suggest a framework and technique for data-driven machine learning that uses satellite remote sensing and aerial LiDAR to forecast plantation stock. This research proposes a novel method in LiDAR image-based carbon emission analysis with their health impact using a machine learning method. Here, the input is collected as LiDAR image-based healthcare data for processing and smoothening. Then, this image has been segmented and classified using convolutional contour graph cut equalization with VGG-net transfer encoder Gaussian neural networks. The experimental analysis for LiDAR image dataset has been carried out for analysis of carbon emission, and healthcare data analysis has been carried out in terms of training accuracy, average precision, recall, F-1 score, and AUC. The proposed technique attained a training accuracy of 98%, an average precision of 93%, a recall of 96%, an F-1 score of 92%, and an AUC of 94%.</p>