This study presents a one-dimensional deep learning model based on convolutional neural networks (CNNs) for predicting the total chlorophyll content of greenhouse lettuce using hyperspectral images. The model incorporates a spectral attention module to dynamically weight the importance of different hyperspectral bands, enhancing the network's representational capacity. The model was evaluated using a five-fold cross-validation on a dataset of 478 images, achieving an average R² of 0.746 and an average RMSE of 2.018. This performance surpasses that of standard methods such as partial least squares regression (PLSR) and random forest (RF), which achieved average R² of 0.703 and 0.682, respectively. The spectral attention module significantly improves the model's ability to discern critical spectral bands, making the deep learning approach more robust and practical for real-world applications. The study highlights the potential of deep learning in non-destructive estimation of chlorophyll content in leafy vegetables, offering a promising tool for high-throughput plant phenotyping and automated monitoring in controlled environments.This study presents a one-dimensional deep learning model based on convolutional neural networks (CNNs) for predicting the total chlorophyll content of greenhouse lettuce using hyperspectral images. The model incorporates a spectral attention module to dynamically weight the importance of different hyperspectral bands, enhancing the network's representational capacity. The model was evaluated using a five-fold cross-validation on a dataset of 478 images, achieving an average R² of 0.746 and an average RMSE of 2.018. This performance surpasses that of standard methods such as partial least squares regression (PLSR) and random forest (RF), which achieved average R² of 0.703 and 0.682, respectively. The spectral attention module significantly improves the model's ability to discern critical spectral bands, making the deep learning approach more robust and practical for real-world applications. The study highlights the potential of deep learning in non-destructive estimation of chlorophyll content in leafy vegetables, offering a promising tool for high-throughput plant phenotyping and automated monitoring in controlled environments.