Abstarct: In reflection seismic, depths near the surface to depth of several in hundred meters are called shallow. shallow depth information for geological engineering studies is very useful, for example, in the making massive structures (dams, large bridges, power plants, industrial structures), mines, deep well drilling (oil, gas and geothermal). Depths less than 100 meters are studied only with refractin seismic methods and surface waveforms and depths of more than a few hundred meters can be studied with deep reflection seismic method. However, there is an operational vacuum in the exploration of structures between about 100 meters.In order to fill this vacuum, shallow reflection seismic was introduced. The difference between this method and the depth reflection method is that the data recording time and sampling distance in the mentioned method are short, and waves with higher frequencies (in contrast to the depth-reflection seismic method) are transmitted to the ground. In this research, Common Reflection Surface and the typical illustration method (NMO/DMO/stack) was used to imagery the shallow reflection seismic data. In the CRS imaging method, the reflected reflectivity of the reflectors in the earth is estimated using the three attributes (normal wavefront curvature, NIP wavefront curvature, central wave entrance angle). In one of the steps of solving the CRS equation in the usual way, the curvature of the normal wave front is assumed to be zero. The mentioned premise for the shallow depths is far from real. To solve the problem mentioned in this study, two new strategies were used to imagery low depth reflection seismic data using the CRS method. The first strategy involves simultaneous simultaneous estimation of the attributes using the PSO algorithm. The second strategy involves estimating the initial values of the attributes in a single-parameter search and a two-parameter search, estimating the final values of the attributes during a three-parameter search around the values obtained from the previous step and its protruding. In all of the second strategy searches, the proposed minimumpim-basis optimization algorithm has been used. The basis of the minimumpim-basis algorithm is the identification and removal of the spaces where the probability of having the optimal points there is very low. For this purpose, the search space is considered as a multi-dimensional space, each of which dimension to one of the variables. At each stage of the repetition of the minimumpim-basis algorithm, the variables in the optimization problem in the search space are selected with the steps of that stage and for the selected values, the value of the fitness function (the function that is supposed to be optimized) is computed and the space around those areas that have less value is removed from the search space. The mentioned methods and the conventional image method (NMO / DMO / stack) were applied to three lines of marine reflection seismic data. The marine data used in this study is Malaysia. The results indicated that the second proposed strategy was more effective than other methods.