In this study, the design for additive manufacturing of shoe molds with complex and precise patterns was performed to achieve rapid prototyping. Low alloy steels such as AISI4340 and SAE1524 were selected to make shoe molds to apply to the conventional chemical etching process. A lattice-oriented design and optimization of toolpath was tested to reduce the processing time. A reduction of 60% in processing time and pattern precision of 0.3 ㎜ was been achieved. Moreover, to improve the reliability of pattern formation, single-layer image analysis with computer vision and machine learning was developed and non-destructive analysis by X-ray CT was been performed. It was found that the quality of shoe molds can be decreased with a single defective layer.