This paper aims to enhance agricultural productivity by detecting nitrogen deficiency (abiotic stress) and rust (biotic stress) in wheat crops. The agriculture sector faces challenges like costly disease treatment, and inadequate facilities. Wheat production faces abiotic stresses, leaf rust, and pests, negatively impacting crop yield, quality, and economic prosperity. Nitrogen boosts wheat yields and protein content, while nitrogen-deficient leaves grow yellow and reduce size, and leaf rust, a fungal disease characterized by dusty, reddish-brown fruiting bodies, grows in moist conditions and 20–25 °C temperatures. The study used deep learning techniques like CNN, GoogLeNet, and DenseNet121 to analyze wheat leaf images for rust and nitrogen deficiency, with DenseNet121 achieving the highest accuracy of 99.37% whereas GoogLeNet and CNN achieved 97.10% and 89% accuracies respectively. DenseNet121 is a powerful method for identifying wheat crop photos related to nitrogen deficiency and rust, efficiently learning from small training data and capturing complex patterns. Implementing and continuously improving these methods could potentially reduce wheat leaf disease, increase agricultural productivity, and ensure food security. Smartphone applications can automate the detection of nitrogen deficiency in wheat leaves and reliably identify leaf rust globally.