具有色偏修正能力之景色分類系統:設計及應用

Abstract

由於網際網路普及，造就數位影像的使用大幅提昇，也帶動影像擷取器的使用風潮，而現今影像擷取器為了讓使用者在不同景色(scene)，都能讓拍攝者完成高品質的拍攝，常會在影像擷取器上提供不同的景色模式(scene mode)，例如風景、海灘…等模式供拍攝者選擇，但此同時也常讓拍攝者在使用上變為更複雜解及不便利。基於此理由，本論文提出具有色偏偵測的單一景色模式使其能適用於不同景色的拍攝，期能解決拍攝者需切換景色模式的不便利性。 本論文提出之景色分類系統，首先會將預覽影像進行影像色偏偵測，若經色偏偵測模組判定影像具有色偏，將依色偏修正模組建議對影像擷取器之參數進行微調，以消除影像色偏之現象產生。此修正後之影像將分割成多個影像單元將各個影像單元的拍攝體的色調、飽和度、亮度、對比度等特徵提取，依據此特徵為基礎對拍攝場景進行分析，最後綜合各個影像單元分類結果來定義出可能之影像場景，然後給于適當之光圈值、快門速度、曝光補償以及ISO感光度等參數之設定。 藉由本論文提出之景色分類系統，拍攝者在拍攝時將無需為細緻的相機設置多費心思，而能夠更加專注於拍攝影像構圖。

With the popularity of internet in recent years, the usage of digital image is growing dramatically, which raises the trend of using image capture device. In order to allow users taking good photos with correct exposure in different scenes, the image capture device provides many scene modes such as scenery, beach…etc for user choosing. However, many scene modes complicate the function of digital cameras and users always feel inconvenient to switch the modes again and again. If a single mode can apply in different scenes, it would solve this inconvenience. The dissertation proposes a Scene Categorization System (SCS) with color cast correction, the preview image is applied into color cast detection, if judged by the color cast detection module with color cast, and the white balance module will depend on the suggestions to fine-tune image parameters to remove the color cast. After that, the image will be split into multiple image block units to extract each image block of the hue, saturation, brightness and contrast value and base on those characteristic values to estimate the shooting scene. Then the image capture device aperture value, shutter speed, and other parameters can be set up automatically. By the scene categorization system, the photographer is not necessary to take too much care with detailed camera settings and will be able to focus on composition.