Résumé
There is no quantitative theory to explain why a high 80% of all planetary nebulae are non-spherical. The Binary Hypothesis states that a companion to the progenitor of a central star of planetary nebula is required to shape nebulae whose shapes are not spherical or mildly elliptical, implying that many single post-AGB stars do not make a PN at all. A way to test this hypothesis is to estimate the binary fraction of central stars of planetary nebula and to compare it with that of the main sequence population. Preliminary results from the infrared excess technique indicate that the binary fraction of central stars of planetary nebula is higher than that of the main sequence, implying that PNe could preferentially form via a binary channel. I will present new results from a search of red and infrared flux excess in an extended sample of central stars of planetary nebula and compare the improved estimate of the PN binary fraction with that of main sequence stars.