An image recognition method for ovulation detection, including: obtaining a pre-identified image； converting the pre-identified image into a gray-scale image； performing a circular detection on the gray-scale image, and obtaining a circular image block； calculating the circle distance between the center of the image block and the four sides of the gray image, and obtaining a distance data group； determining the value of the distance data group and the value of a reference distance value, if the value of the distance data group is greater than or equal to the value referring to the distance value, an image pre-processing step is performed, wherein, the image pre-processing step is used to eliminate the gray-scale image noise； performing a feature extraction step to perform image feature extraction； and performing an ovulation state identification step, wherein, through the ovulation state identification step, determining the pre-identified image belongs to one of an ovulation period image, a non-ovulation period image, or a transition period image. The pre-identification image obtained by the user can be ensured by the ovulation detection image recognition method of the present invention. The method can also increase the recognition success rate to avoid misjudged reading.一種排卵檢測影像辨識方法，包含：取得一預辨識影像；將該預辨識影像轉換成一灰階影像；將該灰階影像進行一圓形偵測，並取得一圓形影像區塊；計算該圓形影像區塊之圓心分別與該灰階影像之四邊的距離並取得一距離數據組；判斷該距離數據組之數值與一參考距離數值之大小，若該距離數據組之數值均大於或等於該參考距離數值，則進行一影像前處理步驟，其中，該影像前處理步驟係用以消除該灰階影像雜訊；執行一特徵擷取步驟來進行影像特徵擷取；及進行一排卵狀態辨識步驟，透過該排卵狀態辨識步驟，判斷該預辨識影像屬於排卵期影像、非排卵期影像或是過渡期影像之其中一種。透過本發明之排卵檢測影像辨識方法可以確保使用者所取得之預辨識影像是合格的影像，可增加辨識成功率避免判斷讀有誤。
