Soil surface roughness (SSR) is an important parameter affecting surface hydrology, erosion, gas exchange and other processes. The surface roughness of the farmland environment is directly related to the tillage process. In order to accurately characterize the random roughness (RR) parameters of the surface after ditching, a three-dimensional (3D) digital model of the surface was obtained by laser scanning under the conditions of an indoor ditching test, and the influence of oriented roughness components formed by removing ridge characteristics on the RR of the surface was analyzed by introducing the wavelet processing method. For this reason, four groups of ditching depths and two types of surface conditions (whether the surface was agglomerated or not) were designed in this paper. By comparing the root mean squared height (RMSH) and correlation length (CL) data calculated before and after wavelet processing under each group of tests, it was concluded that the RMSH values of the four groups before and after wavelet processing all change more than 200%, the change amplitude reached 271.02% under the treatment of 12 cm ditching depth, meanwhile, the average CL value of five cross-sections under each group of ditching depths decreased by 1.43–2.28 times, which proves that the oriented roughness component formed by furrows and ridges has a significant influence on the calculation of RR. By further analyzing the roughness value differences of clods and pits in different directions and local areas before and after wavelet transform, it was shown that the wavelet transform can effectively remove the surface anisotropy characteristics formed in the tillage direction and provide a uniform treatment method for the evaluation of surface RR at different ditching depths.

Soil block distribution is one of the important indexes to evaluate the tillage performance of agricultural machinery. The traditional manual screening methods have the problems of low efficiency and damaging the original surface of the soil. This study proposes a statistical method of farmland soil block distribution based on deep learning. This method combines the adaptive learning rate and squeeze-and-excitation networks channel attention mechanism based on the original Mask-RCNN and uses the improved model to identify, segment and distribute statistics of the farmland soil blocks. Firstly, the influence of different learning rates and an improved Mask-RCNN algorithm model on training results were analyzed. Secondly, the effectiveness of the model in soil block identification and size measurement was analyzed. Finally, the identified soil blocks were classified accordingly, and the scale problem of soil block distribution after removing edge soil blocks was analyzed. The results show that with the decrease of learning rate, the loss value of model training decreases and the prediction accuracy of model is improved. The average precision value of the improved model increased by 25.29 %, and the recall value increased by 8.92%. The correlation coefficient of the maximum diameter measured by manual measurement and the maximum diameter measured by model algorithm was 0.99, which verifies the feasibility of the algorithm model. The prediction error of the model is the smallest when the camera height is 40 cm. Large-scale detection of soil block size in an experimental field in Hefei, Anhui, with an average confidence of over 97%. At the same time, the soil block is effectively classified according to the set classification standard. This study can provide an effective method for the accurate classification of soil block size and can provide a quantitative basis for the control of farmland cultivation intensity.