Medical image processing apparatus of the present invention, compares the feature value calculating section for calculating a feature value for each pixel in the image, the feature of the subject pixel, and a feature quantity of a plurality of pixels located around the target pixel is extracted as a candidate region of pixel group is determined on the basis of the comparison result obtained by a determination unit for determining the target pixel as to whether it belongs to a localized area of the relief structure as belonging to a localized area of the relief structure, topical relief structure The information obtained on the basis of the pixel group of non-candidate regions and information obtained based on the extracted candidate area extraction unit, the pixel group of the candidate region as a non-candidate region pixel group is determined not to belong to the region a candidate area correction unit that uses at least one of within and correcting the extraction result of the candidate region extraction unit, linear structure detection on the basis of the candidate regions which are corrected by the candidate area correction unit detects a region where linear structure I have, and part.本発明の医用画像処理装置は、画像の各画素毎に特徴量を算出する特徴量算出部と、注目画素の特徴量と、注目画素の周辺に位置する複数の画素の特徴量と、を比較した比較結果に基づき、注目画素が凹凸構造の局所領域に属するか否かを判定する判定部と、凹凸構造の局所領域に属すると判定された画素群を候補領域として抽出し、凹凸構造の局所領域に属しないと判定された画素群を非候補領域として抽出する候補領域抽出部と、候補領域の画素群に基づいて取得される情報及び非候補領域の画素群に基づいて取得される情報のうちの少なくとも一方を用い、候補領域抽出部による抽出結果を補正する候補領域補正部と、候補領域補正部により補正された候補領域に基づき、線状構造が存在する領域を検出する線状構造検出部と、を有する。
