Hyperspectral Remote Sensing Inversion and Monitoring of Organic Matter in Black Soil Based on Dynamic Fitness Inertia Weight Particle Swarm Optimization Neural Network

Long-term degradation of black soil has led to reductions in fertility and ecological service functions, which have seriously threatened national food security regional security. This study is motivated by the UN’s Sustainable Development Goal (SDG) 2—Zero Hunger, specifically, SDG 2.4 Food Production Systems. The aim was monitor organic matter (SOM) content its dynamics via hyperspectral remote sensing inversion. great significance effective utilization sustainable development resources. Taking typical area Northeast China as an example, data ground features were compared with SOM contents measured samples correlate spectral features. Based on their quantitative relationship, a dynamic fitness inertia weighted particle swarm optimization (DPSO) algorithm proposed, balances global local search abilities algorithm. DPSO applied parameter adjustment artificial neural network (BPNN), used instead traditional error back propagation algorithm, build DPSO-BPNN model. Then optimal analytical expression inversion obtained improve generalization ability stability results show that model more stable accurate than existing models, such multiple stepwise regression, partial least squares, BP models (adjust complex coefficient determination = 0.89, root mean square 1.58, relative recent deviation 2.93). are basically consistent trend during surface geochemical exploration. As such, this provides basis for monitoring soil.