We have adapted Cu interconnect layers to realize a high sensitivity in a small-pixel CMOS image sensor with a pixel size of 2.5 × 2.5 μm. We used the 1P3M CMOS process, and applied Back End of Line (BEOL) with a design rule equivalent to the 90-nm process. The Cu process features a fill factor that is about 15% greater and an interconnect layer height about 40% less than those of the Al process. As a result, the sensitivity at F5.6 is about 5% greater, while that at F1.2 is about 30% greater. One of problems with the Cu process is the stopper film of the Cu that interferes with the light. Furthermore, this stopper film interacts with the SiO<sub>2</sub> layers to form a multilayer, which leads to a discontinuity in the reflection characteristics at some wavelengths (ripple). Our method involves removing the stopper films together with all the layers. We have also adopted the use of an inner lens. Using these methods, we were able to eliminate the problem of discontinuity in the reflection at some wavelengths. Another problem is the deterioration in the shading characteristics of the optical black area where the black standard is assumed, due to the fact that the Cu interconnect layer is much thinner than the Al interconnect layer. We confirmed that the optical black shading characteristic could be satisfied by using a color filter with the Cu interconnect layer. To realize a 2.5μm pixel CMOS image sensor, we developed the Cu interconnect process. As a result, we created a high-sensitivity CMOS image sensor. This technology should enable the further reduction of the pixel size to less than 2.5 μm.